The next killer app: personalised healthcare
Will our smartphones one day tell us if a heart attack is on its way, and nanosensors in our bodies dispense the medicine to deal with it? Eric Topol thinks so.
Eric Topol isn’t just a top cardiologist at the Scripps Clinic, San Diego, but a terrific man. If you watch the NBC clip of him on how easy it appears to be for a $199 smartphone to display cardiograms, hearts beating or glucose levels, it can make your own heart skip a beat in joy. An early and vindicated critic of Merck’s Vioxx drug, Topol – wiry, super-fit, emphatic – likes to shake things up. His book is not without its flaws, but is an excellent overview of what IT has begun to bring to medicine.
In an early chapter, Topol describes the defects of one-dose-fits-all drugs that produce modest benefits for whole populations of patients. Instead of this ‘population medicine’, he wants drugs that are tailored to individuals’ genetic characteristics – personalised medicine. He makes some familiar points about the false positives and side effects that surround mass, blunt, untailored screenings for prostate cancer (the prostate-specific antigen (PSA) blood test) and for breast cancer (mammography). Like others, he worries that the findings from conventional large-scale clinical trials are less likely to be true if the trials in question are small, the benefits small, the associated financial and other interests great, and the scientific field ‘hot’. His purpose in all this is a properly modernising one, in that he believes that the judicious use of IT can help bring out ‘real evidence based on individuals, not populations… an era characterised by the right drug, the right dose, and the right screen for the right patients, with the right doctor, at the right cost’.
Some of the author’s enthusiasm for IT in medicine would, it seems, have a bit to do with American individualism; but Topol turns in a notably balanced chapter on IT-assisted advances in sequencing the human genome. In 1997, it was only possible to identify one, very personal, single nucleotide polymorphism (SNP) at a time; by 2007, chips and automated robotic systems allowed a million SNPs to be tracked in an individual. In the middle of that year, a genomics gold rush began, with a gene governing susceptibility to a particular disease discovered nearly every week. Yet as Topol notes, snapshots of genetic variations that last a lifetime don’t pinpoint the mechanism for susceptibility to disease – the figures for which are anyway still based on populations, and not on individuals.
Topol notes the negative comments that accompanied the tenth anniversary, in June 2010, of the first draft human-genome sequence. What the complaints about lack of progress missed, he convincingly argues, is the very pragmatic implications of one branch of genome-wide association studies (GWAS). When, in 2005, the first GWAS was published, the link it drew between a particular genetic variation and age-related macular degeneration was a bit of a lucky strike: in general, many variations typically exist for each disease, and their interaction, in networks, is a subject in its own right – systems biology. But pharmacogenomics, in which GWAS track the key genes involved in a prescription drug’s therapeutic action and how genetic variants can ensure side effects, has much more immediate promise.
By contrast with GWAS, in whole-genome sequencing, digital maps of six billion letters apiece are made fast, cheaply, but with limited accuracy. Comprehensive data about an individual can be got for perhaps $1,000 – but interpretation costs several hundred thousand dollars. When a sequence is annotated by, say, 30 authors, several hundred hours of work is involved. Altogether, data analytics, computational biology and medical informatics make the process a costly one, which is why clinical applications of WGS have been limited mainly to cancer therapies (determining the importance of protein kinase genes, for example). Still, Topol holds out hope of mankind developing the ability to sequence the DNA of the tumour of any individual afflicted with one.
The book is very clear about the benefits that wireless sensors can offer patients. Wireless healthcare, beginning some years back as a fitness-related pursuit (think Nike and wireless accelerometers recording data as people move), has moved on to continuous glucose-monitoring (CGM), in which a dedicated transmitter is inserted with a 27-gauge needle just below the skin of the stomach. Yet even here there are limitations: the integration of CGM with insulin pumps, in closed loops, has yet to be achieved. Still, the convenience, and continuous nature of the monitoring that is now possible is only partly qualified by the fact that most diagnoses based on such monitoring have yet to be automated. The next logical step, Topol suggests, will come when monitoring services help patients manage chronic disease.
Underestimating the impediments to innovation
By the time he puts together wireless sensors with genomics, Topol has an amazing story to tell. Ideally, he says, the monitoring of hearts, for example, would work with implanted nanosensors, smaller than a grain of sand, which would communicate with patients’ smartphones – and ‘individuals who would get the nanosensors would be those whose genome sequence or other biomarkers had already put them at risk for heart attack’. Altogether, this would allow ‘real prevention for the first time in medical history’. One day, indeed, ‘nanosensors may be able to release medications on their own in response to high levels of circulating cells or nucleic acids’.
Topol also catches the imagination by showing how different methods of forming images of any part of the body may eventually lead to the ability to print bespoke organs. Readers of spiked will know, however, about the excessive claims for IT made by American enthusiasts for 3D printing, and it’s all too clear that Topol writes very much in the tradition of upbeat technological determinism long ago established by Wired magazine. He is convinced that IT has begun the ‘democratisation of DNA’. And in this, he underestimates the impediments to innovation in medical IT.
Topol rightly notes that medical reform in the US has been much more about improving access and insurance coverage than it is about innovation. Yet he frets about IT being responsible for the depersonalisation of patients, and about it raising concerns over privacy and data being sold to pharmaceutical companies. Despairing of doctors, the life-science industry, government and health insurers, he believes that it is consumers, and consumer rights, that will finally bring powerful mobile apps – along with imaging and patient records – to the field of medicine.
This consumer worldview makes not just for over-familiar ideas about recent developments in IT, but for poor politics. We learn once more about how netizens, digerati and ‘digital natives’, this time ‘as defined by people under age 30’, will own the future. On the other hand, Topol is insouciant about proclaiming that IT can lead to ‘a sense of disconnectedness reflecting addictive behaviour’. He goes on: ‘Although it remains controversial whether the internet and connectivity directly lead to adverse anatomical and functional brain effects, few would debate whether it [sic] has affected our behaviour.’ His sources, in these complicated matters, are more popular than one might otherwise anticipate.
Describing how a wrist transceiver can allow arterial blood pressure to be detected, with only the occasional calibration with an arm cuff needed, Topol adds: ‘It doesn’t take too much imagination to figure out that this is a game changer in health monitoring – or that it might create a legion of e-hypochondriacs… While many individuals who surf the web start believing they may have the diseases they are reading about, continuous monitoring of one’s vital signs takes this concern to a new level. Nevertheless, the upside of having such information available is considerable.’
In even-handedly logging both every ‘upside’ and every ‘downside’ of IT in medicine, Topol passes over the structural barriers standing in the way of IT’s more rapid diffusion in the discipline. The very individualism that he so often invokes – now articulated in the therapeutic language of ‘rights’ in general and the right to personal medical information in particular – tends to be premised on fear: the fear of not knowing the whole truth about one’s body. It is also founded on the desire for the state to regulate such matters. Even though IT is late to the party, therefore, healthcare is already a highly individuated business, in which each of us all too easily rushes to a worst-case scenario that has usually first been aired in an official press release.
In this atmosphere, breakthroughs in medical IT are likely always to meet up with very substantial barriers – barriers not of a technical, but of a political character. Has Britain’s drive to build a national system of electronic patient records proved an expensive disaster just because of private-supplier greed? Or did it happen, as is more likely, because of the scale of bureaucratic gold-plating that accompanied the project, intent on excluding all risks? In his treatment of social media and medicine, Topol observes that ‘in these highly interactive online sites, consumers are teaching each other. This feeds the lack of trust in the medical profession… but it has positive attributes as well… [and] can provide extraordinary emotional support’. That is too rosy a judgement. To the extent that it has helped some patients, today’s IT has also disorientated others, and, regrettably, has tended to dethrone medical expertise.
Altogether, IT doesn’t just help solve medical and technical problems. In today’s culture, and particularly because it is allied to health issues, medical IT is itself perceived as, and is often in actuality turned into, a powerful source of new problems, and especially the problem of who to trust for healthcare advice.
Altogether, Topol is a better doctor than he is a sociologist. Read his fascinating book with this in mind, and prepare for what promises to be a bare-knuckle fight: the fight for more and better medical IT.
James Woudhuysen is professor of forecasting and innovation at De Montfort University, Leicester, and editor of Big Potatoes: the London manifesto for innovation.