Pharmacists in Australian general practice: an opportunity for expertise in precision medicine

At the end of May the Australian Medical Association (AMA) released a proposal for the Australian Commonwealth Government to establish a funding model to integrate nondispensing pharmacists into general practice – the Pharmacists in General Practice Incentive Programme (PGPIP) [Australian Medical Association, 2015]. This is built on joint work by the AMA and the Pharmaceutical Society of Australia. A similar collaboration exists in the UK between the Royal College of General Practitioners and the Royal Pharmaceutical Society, with a joint statement promoting the role of pharmacists in general practitioner (GP) surgeries released this February [Royal College of General Practitioners and Royal Pharmaceutical Society, 2015]. The Australian PGPIP proposal nicely summarizes the evidence to support improved quality use of medicines (QUM) and decreased adverse drug events when pharmacists work directly alongside GPs. Independent economic analysis showed cost savings of $A545 million to the Australian healthcare system over 4 years, primarily through fewer avoidable hospital admissions and a reduction in the utilization of medicines [Deloitte Access Economics, 2015].

The PGPIP funding proposal is a very important step in expanding the role of pharmacists in Australian primary care. ‘Practice pharmacists’ would become commonplace in GP clinics, like practice nurses, and would focus exclusively on high quality medicines management. As outlined in the proposal, nondispensing pharmacists in general practice would:

1. conduct medication management reviews

2. give patients medication advice to increase compliance and medication optimization

3. support GP prescribing

4. liaise with outreach services and hospitals when patients are discharged on complex medicine regimens

5. update GPs on new drugs

6. conduct quality and medication safety audits

7. develop and manage drug-safety monitoring systems [Australian Medical Association, 2015].

In addition to these activities, we believe that pharmacists in general practice could become experts on the clinical utility of ‘precision medicine’, in particular, new technologies designed to improve drug efficacy and safety.

Precision medicine has recently been defined as ‘treatments targeted to the needs of individual patients on the basis of genetic, biomarker, phenotypic or psychosocial characteristics that distinguish a given patient from other patients with similar clinical presentations’ [Jameson and Longo, 2015]. This concept is not new, but rapid advances in medical science, particularly in oncology with the molecular classification of solid tumours [Gillis et al. 2014], have accelerated research efforts to realize this concept more broadly in clinical practice. Importantly, next-generation DNA-sequencing costs are decreasing to a point that in the near future it will be affordable to sequence the whole genome and store it in an electronic health record, thus allowing real-time access to a wealth of information during consultations. In the State of the Union address in January, President Obama launched ‘a new Precision Medicine Initiative to bring us closer to curing diseases like cancer and diabetes’, pledging $US215 million to the National Institutes of Health to accelerate progress towards this new era (www.whitehouse.gov/sotu) [Collins and Varmus, 2015]. Top universities are embracing this too. Columbia University has made precision medicine an institution-wide research priority (http://newsroom.cumc.columbia.edu/precision-medicine/).

This research will generate an avalanche of new information to characterize patients and disease, proteomic, metabolomic, genomic, etc. There will be many new technologies, not yet invented, claiming to benefit human health. An area of research aimed at improving therapeutics and QUM is pharmacometrics. Pharmacometrics is ‘the science of developing and applying mathematical and statistical methods to characterize, understand and predict a drug’s pharmacokinetic and pharmacodynamic behaviour’ [Perera et al. 2014]. Put very simply, this is the use of computers to guide dosing. This approach has been applied successfully in the pharmaceutical industry to facilitate drug development but has had limited use in the clinic [Jones et al. 2015]. Notable exceptions include the individualization of aminoglycoside dose [Phillips et al. 2015] and the dosing of antibiotics in critically ill patients [Roberts, 2011]. Within this field, it is now possible to create virtual humans using physiologically based pharmacokinetic (PBPK) modelling and simulation software [Jones and Rowland-Yeo, 2013]. The characteristics of a patient known to influence pharmacokinetics and pharmacodynamics, that is, a patient’s age, weight, liver size, renal function, genetics, etc., can be entered into the software to ‘individualize’ models based on population data. Simulations are then run to determine which drug or dose is best in the ‘computer twin’, and, in principle, your patient.

This approach is more powerful than pharmacogenomics alone because it allows greater sources of variability in drug response, including genetics, to be considered simultaneously: for example, a patient (VKORC1 wild type and CYP2C9 poor metabolizer) with alcoholic chronic liver disease and newly diagnosed nonvalvular atrial fibrillation who is about to commence warfarin and is taking simvastatin, metoprolol and fluvoxamine, a strong inhibitor of warfarin metabolism [Polasek et al. 2011]. An often overlooked limitation of pharmacogenomics is that genotype is fixed throughout a person’s lifetime, but in reality there are many varying factors that influence phenotype, including age, comorbidities, cigarette smoking and exposure to interacting drugs and/or chemicals. The PBPK approach is able to consider all these when making predictions. In the above example, simulations could be re-run to pre-empt any changes in warfarin dose when it is decided to cease fluvoxamine (i.e. removing the inhibition of warfarin metabolism), thus decreasing the risk of under-anticoagulation and thromboembolic stroke. The relatively poor success of personalized medicine in the past, which has been based primarily on patient genotype, may not be indicative of its future potential.

PBPK modelling and simulation is only one example of technology currently being developed for precision medicine. Such technologies are complex and will require a considerable breadth of understanding and skill to evaluate potential clinical utility. In a precision-medicine era, GPs will have their hands full navigating referral pathways with many more branches as a result of improved diagnostics, that is, connecting patients with a specialist who has access to the emerging data and clinical guidelines [Jameson and Longo, 2015]. In this setting, pharmacists using technologies that genuinely improve QUM will be highly valued, and could become the hub of expertise in this field for the healthcare sector.

In conclusion, the PGPIP is an important proposal for pharmacy in Australia – a welcome expansion of the pharmacists’ role in primary care. This is consistent with a similar move in the UK. Focusing exclusively on medicines management, nondispensing pharmacists in general practice should aspire to expertise in emerging technologies designed to improve QUM. This will require considerable training and commitment to ongoing education, but would offer a challenging and rewarding career.