In this large prospective cohort study of men with prostate cancer receiving ADT, we report a high baseline prevalence of cardiovascular risk factors, and of reduced bone mass even prior to commencement of ADT. After 2 years of continuous ADT, men experienced central weight gain and, unless treated for pre-existing diabetes, worsening dysglycaemia. Despite this, reductions in systolic and diastolic blood pressure, total cholesterol and LDL were observed with treatment. In addition, while achieving vitamin D sufficiency, a significant decrease in BMD occurred, unless anti-resorptive therapy was given.
Obesity, insulin resistance and cardiovascular risk
The 87% baseline prevalence of being overweight or obese in our prostate cancer cohort is higher than the 73% prevalence rate in age-matched men in the general Australian population (Cameron et al., 2003). Similarly hypertension, hypercholesterolaemia, impaired fasting glycaemia, diabetes and pre-existing cardiovascular disease was more common in our cohort than in age-matched Australian men (Dunstan et al., 2002). This is consistent with observations that obesity is a risk factor for prostate cancer as well as for cardiovascular disease (Grossmann & Wittert, 2012). The increase in central adiposity during ADT in the absence of increases in body weight is consistent with findings in smaller prospective cohort studies, which showed that while ADT increases central fat, it also led to a loss of muscle mass, a phenotype which has been termed sarcopaenic obesity (Basaria et al., 2002; Smith, 2004; Hamilton et al., 2011). Therefore, reliance on body weight alone may underestimate the deleterious changes in body composition that occur during ADT. This study suggests that central weight gain is a dominant feature of ADT, since it developed despite diet and lifestyle advice, and we found no predictors of an increase in waist circumference to guide targeting of high-risk groups.
In the subgroup of men without diabetes, HbA1c levels increased significantly, likely due to increased insulin resistance as a consequence of central weight gain. Although we did not measure indices of insulin resistance, previous smaller, short-term studies showed that insulin resistance increases within months of ADT commencement (Smith et al., 2008; Hamilton et al., 2011). Our findings of worsening glycaemia after 2 years of ADT are also consistent with findings of large, retrospective registry studies that have associated ADT with a 15–45% increased relative risk of diabetes, with estimated numbers needed to cause one case of incident diabetes ranging from 15 to 90, for review see (Grossmann & Zajac, 2011a). In contrast, in our men with pre-existing diabetes, HbA1c remained stable, suggesting that proactive management can prevent a deterioration of glycaemia during ADT. This observation is especially relevant in the light of recent findings from an observational database study showing that ADT was associated with an increase in HbA1c despite intensification of anti-diabetic therapy (Keating et al., 2013).
Previous studies without lipid-lowering intervention have shown that during the first year of ADT, in line with the increased central adiposity and insulin resistance, levels of triglyceride increase by more than 25%, and total cholesterol by up to 10%. LDL cholesterol levels are variable (from no change to +7%), and HDL increases by 8–20% (Grossmann & Zajac, 2011b). Thus, increase in HDL levels just short of significance (p = 0.060) seen in our cohort despite central adiposity are consistent with these previous short-term effects of ADT and indicate that these theoretically cardioprotective changes are maintained in the longer term. Interestingly, triglyceride levels remain stable over 2 years, which may be a result of diet and lifestyle advice provided. In addition, men treated in the clinic according to standardized guidelines had a significant reduction in mean total cholesterol and LDL. Furthermore, because of treatment, blood pressure was effectively lowered in our cohort. Thus, proactive management combining lifestyle advice and pharmacotherapy significantly reduced blood pressure and LDL cholesterol, arguably among the most dominant modifiable cardiovascular risk factors. This is pertinent given that, although somewhat controversial (Nguyen et al., 2011), ADT has been associated with increased cardiovascular morbidity and mortality (Basaria, 2008; Grossmann & Zajac, 2011a). Given that most men with prostate cancer do not die from their disease, but instead from cardiovascular causes, even moderate ADT-associated increases in cardiovascular risk factors, may, if unmanaged, offset some or all of the benefits of this therapy. In our cohort, incident cardiovascular events were few, with three new diagnoses of ischaemic heart disease and one diagnosis of cerebrovascular disease among the 113 men. Larger controlled studies will be required to assess the impact of risk factor modification on cardiovascular outcomes in men receiving ADT.
Fifty per cent of the men in our cohort presented with reduced bone mass even before ADT was commenced, and the majority of the cases were undiagnosed prior to attendance at the clinic. This is consistent with observations that assessment of bone mass is underused in older men in the general population as well as in men with prostate cancer receiving ADT; in a recent SEER database analysis, only 10% of men had documented bone density testing in the 18-month period beginning 6 months before ADT initiation (Morgans et al., 2013). Men receiving ADT experience accelerated bone loss, up to 11% during the first year of ADT, and this may be underestimated by conventional DEXA (Hamilton et al., 2010). Identification of individuals with osteoporosis or high risk of fracture is imperative, given that a large population-based study has shown that ADT increases 5-year absolute fracture risk by 7% (Shahinian et al., 2005). In this study, the number needed to harm for ADT to cause one fracture was estimated to be 16–28. In addition, men who sustain a fracture have a two to threefold increased risk of mortality (Beebe-Dimmer et al., 2012). The ability to institute effective anti-resorptive therapy with bisphosphonates or denosumab which prevent bone loss and decrease fracture risk (Greenspan et al., 2007; Smith et al., 2009) makes targeting this group imperative.
In our cohort, those identified with osteoporosis and who were treated, maintained their lumbar spine and total hip BMD. Not unexpectedly, those who were untreated had a decline in both total hip and lumbar spine BMD over 2 years which was consistent with previous reports (Greenspan et al., 2005). Interestingly, those with higher baseline total hip BMD had greater relative loss of BMD, indicating that a higher bone density may not protect against ADT-induced bone loss. This re-emphasizes that all patients receiving ADT should undergo periodic assessments of bone health and receive treatment as appropriate (Grossmann et al., 2011).
In summary, in this large prospective cohort of men with prostate cancer with long-term follow-up while attending a dedicated service, we find a high baseline prevalence of cardiometabolic risk factors and reduced bone mass, higher than expected from the general population. Despite lifestyle advice and vitamin D supplementation, men experience increased central adiposity, and, unless pharmacotherapy is instituted, worsening glycaemia and further reductions in bone mass. However, with the caveat that this was an observational study, our results also indicate that proactive management can reduce cardiovascular risk factors and prevent bone loss. The proportion of men achieving targets such as blood pressure, LDL cholesterol, vitamin D sufficiency and anti-resorptive therapy (Table 1) increased from one third at baseline to more than half at the 2 year follow-up. It should be noted that this was not a randomized trial in a population selected to be eligible for these interventions. Reasons for not achieving targets included contraindications to treatment, non-adherence or refusal, or ongoing optimization of therapy at the last follow-up.
A limitation of this study was the absence of a control group, as the vast majority of our men had high-risk non-metastatic prostate cancer, where long-term ADT adjuvant to definitive radiotherapy is associated with clear mortality benefits (Bolla et al., 2009). In addition, due to short follow-up and limited number of events, this study focused on risk factors, but is limited by lack of data on cardiovascular and bone decay outcomes, and survival. Arguably, failure to prevent central weight gain was a limitation of this study, because body composition changes are considered important drivers of ADT-associated cardiometabolic risk, and may contribute to fracture risk (Grossmann & Zajac, 2011a). Implementation of an intensive lifestyle programme was beyond the scope of our clinic. However, there is evidence that low testosterone increases insulin resistance independent of changes in body composition (Grossmann, 2011). In addition, testosterone may affect cardiovascular risk via direct vascular and myocardial effects (Yeap, 2010). We did not address strategies to mitigate constitutional and sexual side effects of ADT, given the absence of evidence-based management guidelines and of well-standardized outcome measures. However, we did confirm previous reports (Grossmann & Zajac, 2012) that ADT was associated with a significantly increased incidence of anaemia. Indeed, prevalence of anaemia more than doubled, and older men were particularly susceptible. Finally, we did not measure patient-important factors such as engagement in lifestyle recommendations or adherence to medications. Therefore, we do not know the extent to which lack of engagement may have limited the benefit achieved.