Over the past few months, pharmaceutical companies have been dominating the front pages of the business press as the industry experiences a splurge of M&A activity. The deal grabbing most attention is Pfizer’s $152-billion tie-up with Allergan, a firm best known for manufacturing botox. The deal is structured as an inversion, which means that even though Pfizer is putting up the money, it can technically be treated as the acquiree as it will be moving its HQ to Dublin, where Allergan is domiciled, allowing it to benefit from the greatly reduced tax rate that being based in Ireland brings.

Whatever you think of this from an ethical standpoint, it’s clear that in an age where the average drug costs $2.6 billion to be brought to market, a figure that will only rise as the role of biologics increases, pharma companies are looking to cut costs in whatever way they can.

Probably the main area where costs can be reduced, through the creation of new products and methodologies, is clinical trials. In 2012, a team led by Avik Roy of the Manhattan Institute think tank examined drug development in four key areas
of public health. It found that 90% or more of development costs were incurred in phase-III trials, the definitive test where the drug is trialled on a large group of people and decreed to be more or less effective than the existing gold-standard treatment.


An area that has shown promise is the use of trial data derived from point-of-care (POC) testing. This is usefully defined by the College of American Pathologists as "tests designed to be used at or near the site where the patient is located, that do not require permanent dedicated space and that are performed outside the physical facilities of the clinical laboratories."

While randomised, multisite clinical trials are undoubtedly the best measure of a treatment’s efficacy, they take a long time – particularly in the case of chronic illness – are highly expensive, and lessons learned take a while to be translated into clinical practice. POC testing is much cheaper and produces results more quickly, but there have long been questions over its use in the clinical trials space, whether as an indicator of which treatment to administer or as a generator of trial data.

Many of the obstacles to the use of POC testing in trials are technical in nature. According to ‘Point-of-Care Testing Needs’, a 2010 paper by Christopher Price, Andrew St John and Larry J Kricka, any POC testing device must possess a number of qualities. In potted terms, it must be simple to use, safe, employ robust and easy-to-store reagents and consumables, and be accurate enough that the results it produces are in line with existing laboratory tests. Additional requirements are that it provides a quantitative result, offers some kind of decision-making support and, ideally, has the option of connectivity to the patient’s health record or another information system – all in a manner that makes cost sense to health services and insurers.

Unsurprisingly, ticking all of these boxes has not been easy, although according to Jorge Villacian, chief medical officer at Janssen Research and Development, which is owned by Johnson & Johnson, technological developments have started to provide answers. Devices that accurately test HbA1C, PTINR, haemoglobin and cholesterol, among other conditions, are making a real difference on the market.

Ultimately, researchers need to demonstrate without doubt that using data generated at the point of care has a direct impact on clinical outcomes.

"I think point of care is becoming more mainstream as the new technologies become available in different settings," he says. "Probably the best example is for infectious diseases, where people need to make decisions relatively quickly and those decisions depend on the information they get from diagnostic tests. In the past, most of the time, you had to wait for the identification of organisms through different culture methods. Nowadays, more new products are coming out on the market giving information that is needed.

"We are not yet at the stage where we can replace the traditional methodologies for identifying bacteria or viruses but, for some diseases, it’s getting there. We are able to get information quickly, almost at the bedside, and act upon it," he adds.

In recent years, Janssen has formed a number of partnerships with device manufacturers in an attempt to catalyse the development and journey to market of POC devices. Its most recent, announced in March 2015, is with the medical equipment arm of Royal Philips, under which the two will work towards the development of a handheld
blood-testing tool for patients with neuropsychiatric disorders, with Philips responsible for the development and manufacturing of the device, and Janssen taking care of the clinical studies and eventual commercialisation.

The blood test has a head start in that it is based on Philips’ existing Minicare I-20 system for immunoassays. Consisting of a handheld analyser, proprietary software and a disposable cartridge containing the relevant test, the aim is, according to a statement by Janssen, for the device to be able to "detect multiple target molecules at low concentrations within the same blood sample and to show the results on the analyser within minutes".

"We are at the level of designing the assay and testing the analytical performance of the test," Villacian says. "I think the next stage is to take it into clinical samples and do the trials. We are doing the analytical validation of the test and will enter into clinical validation pretty soon."

Test conditions

POC data has been used in clinical trials on a number of occasions in the past decade, with a relatively recent study, developed by Stanford University biostatician Philip Lavori and a team from the Veterans Affairs (VA) Boston Healthcare System in 2011, a particularly interesting case. The aim was to compare drugs and procedures that were already in use by combining the analysis of POC data with the randomisation of a traditional clinical trial, removing the selection bias associated with observational tests.

This particular trial was designed to test weight-based determinations of insulin dose against sliding-scale determinations in non-intensive-care unit patients with diabetes. The primary end point was the length of stay in hospital, and the secondary was glycemic control and readmission within 30 days. On entering the hospital, the point of care, patients are given three options for how to take their insulin: options two and three are weight-based and sliding scale respectively, while option one is ‘no preference’, which leads to an invitation by the clinician to enrol the patient in the study.

If the patient agrees to participate, they are randomised to either of the two treatments and the nurse enters a progress note about the study into the electronic record. The computer system from then on writes orders for the patient and tracks the progress of the respective treatments to see which is providing the best outcome. As the sample size grows and one clear outcome starts to emerge, the most successful treatment becomes standard practice before the clinical trial has even finished.

"The first implementation of a POC test in the VA has demonstrated the feasibility of this new method of evidence production," said the team in the Journal of the American Medical Informatics Association. "Existing functionalities within the VA’s EMR system are currently employed to identify eligible patients, facilitate enrolment, perform randomisation and collect longitudinal data. Early results show patient and clinician acceptance of the integration of a clinical trial into routine clinical care, although more work needs to be done to understand stakeholders’ perspectives."

As with so many things in medicine, the human element is the most important, and often trickiest, to negotiate. Once the role of a clinician starts to blur with that of a researcher, it can affect a doctor’s relationship with the patient and the accepted rules upon which it is based. For example, how will clinicians react to having to randomise the treatments they assign and will this affect the guidelines they must follow? Equally, how will patients react?

According to a group of researchers from the US Board on Health Sciences Policy at the Institute of Medicine writing in the 2012 book Public Engagement and Clinical Trials: New Models and Disruptive Technologies, this is exacerbated by a lack of understanding between clinicians and researchers from the early stages. There is a "lack of appreciation and reward for collaborative work within academic medicine" that needs to be overcome to be more in keeping with the real clinical environment, which is increasingly defined by multidisciplinary work.

The book continues: "The days of lone investigators owning data and carrying out projects in isolation are numbered… Yet, the academic infrastructure has not even begun to dismantle these silos. Additionally, Zuckerman [professor emeritus at University College London] mentioned that the training environment needs to change, so that medical schools produce physicians who are clinical trialists and clinical research courses become a standard part of the curriculum."

Burden of proof

Ultimately, researchers need to demonstrate without doubt that using data generated at the point of care has a direct impact on clinical outcomes and that rapid bedside testing can be as or more effective than existing methods.

The burden of proof is extremely high, even more so than it was in the past. Researchers need to demonstrate, for example, that data taken from POC tests from a patient with a respiratory infection is a good substitute for actually analysing the growth of bacteria in a culture medium. This will require many successful trials and many thousands of bytes of data.

"I think, in the past, a lot of these things that are today the standard of care, we did not subject to the level of evidence generation that we do [with] the new technologies," Villacian says. "I would say, in a way, that new technologies are at a disadvantage because now scrutiny is much higher than it was 15, 16 or 17 years ago, when there was no alternative and you could only go with what was the best thing. These things became ingrained in clinical practice just because that was the only thing available. Today is a different world, and people and the payers as well are trained to think differently."

Despite these obstacles, the success of the trials already carried out and the pressing need to reduce costs bode well for the future of trials that employ POC data.