Chronic effects of hyperprolactinaemia
Osteoporosis
Studies of people with prolactinoma have demonstrated that chronic hyperprolactinaemia is associated with reduced bone mineral density (BMD) and a 2-4.5 fold increased risk of osteoporotic fracture [36,37]. The degree of osteopaenia is related to hyperprolactinaemia duration, especially where the onset was during adolescence, rather than absolute prolactin concentration [38]; treatment of the prolactinoma improves BMD [39].
Longstanding gonadal axis suppression is the most important mechanism linking hyperprolactinaemia with reduced BMD [40] but prolactin may also have a direct inhibitory effect upon osteoblasts, at least in rats [41]. The molecular mechanisms underlying these direct effects are not fully understood but may involve activation of nuclear factor κ B ligand (RANKL), an important regulator of osteoclast differentiation and function [42].
Fractures occur 1.2 - 2.5 fold more frequently and at a younger age in people with SMI [43-46]. The underlying reason for this increase is multifactorial as risk factors for reduced BMD are common in people with SMI. Nevertheless, antipsychotic treatment probably contributes to this risk. A UK study found that hip fracture occurred 1.73 fold more commonly in people with a diagnosis of schizophrenia but 2.6 fold more commonly in those prescribed an antipsychotic [43]. Similarly a Dutch study found a 1.68 and 1.33 fold increased risk for hip fracture for current and past users of antipsychotic drugs respectively. FGAs but not SGAs were associated with increased fracture risk [46], possibly explained by a lower mean BMD in people receiving FGAs and risperidone compared with olanzapine [47,48].
Prolactinoma
An increased prevalence of prolactinoma has been reported in people receiving risperidone and haloperidol, but not ziprasidone, olanzapine, clozapine or quetiapine [49,50]. To what extent this reflects a true increase in incidence is unclear as these data are almost certainly subject to surveillance and reporting bias [51]; people receiving risperidone and haloperidol appear more likely to receive prolactin screening, more likely to obtain a diagnosis of hyperprolactinaemia and more likely to have pituitary imaging than people receiving other antipsychotics. When a tumour is then found, clinicians are more likely to report this because of its rarity.
There are a few case reports of enhanced prolactinoma growth in people treated with antipsychotics [52-54] but it is unclear whether this is related to treatment as prolactinoma growth rates are variable between individuals. While the human epidemiological data are of uncertain significance, D2 receptor blockade by antipsychotics in animals induces lactotroph proliferation and transgenic mice with D2 receptor deletions develop large prolactinomas [55]. While caution is needed in extrapolating these data to humans, there is a theoretical basis for concern and further vigilance is needed.
Cancer
Although a controversial area, recent human epidemiology and molecular biology findings suggest that increasing serum prolactin concentrations may be associated with human breast cancer, and possibly prostate cancer [56]. A detailed description of this debate is beyond this review’s scope and so the subsequent discussion focuses on studies of people with SMI and in particular studies of the effects of antipsychotics.
A recent meta-analysis of six studies found a 12% increased risk of breast cancer in people with schizophrenia compared with the general population [57]. While this study cannot prove causality or determine what aspect of schizophrenia conveys the risk, it was suggested that this may partly result from antipsychotic induced hyperprolactinaemia [57]. There are surprisingly few studies specifically examining the association between antipsychotics and cancer. A US study comparing 52,819 women who had received dopamine antagonists (both antipsychotics and anti-emetics) with 55,289 controls found a 16% increased risk of breast cancer in those who had received treatment [58]. Amongst women with prolonged treatment (>6 years), the risk was increased 2.4 fold. Although the power was limited, it is noteworthy that breast cancer risk was increased in those taking phenothiazines (e.g. chlorpromazine, perphenazine) but not butyrophenones (e.g. haloperidol), despite both classes increasing prolactin by a similar amount. A Danish study
of a population-based cohort of 25,264 antipsychotic users, however, found no increase in the risk of breast or prostate cancer [59]. Further research is needed to assess the extent, if any, to which antipsychotics contribute to the increase in breast cancer in women with SMI.