It was once said that speculation about the future is dangerous, and saying ‘yes’ when asked to speculate about the future is probably more so. However, with a persuasive new Editor of The Bulletin needing copy, I hope that I may be forgiven!
I was probably an easy target, as I have had the privilege of leading the College’s Inter-Specialty Committee for Molecular Pathology (ISCMP) during its first four years. When I agreed to start the Committee, I was told that pathology was on its way out and that molecular methods, especially DNA sequencing, would take over from other outdated methods such as those used by histopathologists or microbiologists. It has been an interesting journey, with quite a few twists, but I continue to believe that molecular pathology is the glue that will hold our various specialties together.
The ISCMP has been busy. Amongst other things, we have cooperated with industry, charities and other organisations to produce guidance for laboratories undertaking cancer molecular pathology,1 produced the new FRCPath curriculum in molecular pathology, and written the CMD ImPACT tools for business planning (www.rcpath.org/cmd-impact.html).
Rumours of the death of pathology as we know it as a result of advances in genetics are undoubtedly premature. Those who believe that any of the 19 pathology disciplines will become unnecessary as everything is increasingly explained by genetics alone are likely to be disappointed, and busy geneticists will undoubtedly be very relieved. That said, there is little doubt that all pathology specialties and pathologists are changing more rapidly than at any time since Virchow developed theories of the cellular basis of disease. The understanding of disease at a molecular level continues to advance, though a cellular understanding of disease is still as necessary as ever – whether to understand the results of a full blood count or the process of cancer metastasis. Cells do, of course, alter their behaviour largely as a result of their molecular makeup, but genetics and mass spectroscopy are unlikely to replace histopathological examination of tissue, which allows the organisation of tissues, cells and molecules to be examined in increasingly greater detail.
The importance of training in pathology
Pathology is first and foremost a scientific discipline. The study of disease underpins medical practice and those who fail to learn the lessons of pathology are likely to rue their lack. Medical schools should take note, and the RCPath’s recently published medical undergraduate curriculum is an important step to ensure that all doctors have an adequate basis for their subsequent practice. More than ever before, medical students need to understand the causes and mechanisms underlying disease to prepare for a world in which a scientific understanding of medicine is essential to prevent disease, make diagnoses and manage patients correctly.
The rapid pace of advance means that it is not enough simply to educate the next generation of medical and scientific staff in pathology. We need to engage and inform the current generation, and in turn train those outside pathology to use the reports we produce to benefit patients. Pathologists of many specialties may end up leading multidisciplinary teams, challenging and explaining management options for patients, rather than simply giving reports.
The clinical role of pathologists now goes far beyond diagnosis. Whether we like it or not, we are increasingly involved in decisions on treatment, and monitoring of response to treatment is a major role for blood sciences in particular. Understanding of aetiology and pathogenesis must therefore be matched by an understanding of what treatment will do at every level – molecular, cellular, whole organism and whole person.
We need to personalise treatment because we are all different. We differ genetically, the bacteria and viruses that infect us differ, our lifestyles differ and the environments in which we live differ. Those differences mean that we respond differently to different treatments for disease. For instance, analysis of the individual make up of cancers using cells, protein, RNA and DNA allows us to predict their behaviour in response to particular treatments. When I first started doing this, in 1990, I was pretty universally told that I was wasting my time. For many cancers, personalised therapy based on laboratory diagnosis is now routine practice and is producing amazing results for patients. We now have the ability to tailor treatment to individuals, with targeted drugs and companion diagnostic tests. I still prefer the term ‘personalised medicine’, though stratified, precision, or individualised medicine are the same thing.2
The methods used to provide personalised medicine are evolving rapidly. There are now many DNA sequencing and other methods available. The choice of which one to use depends on many factors, particularly the diagnostic information required and when it is required. Timely diagnosis is essential for many patients, and should surely influence when and where tests are performed. Whole genome sequencing may be cost effective for genetics, where results may not be used for years, but not for molecular pathology, where rapid diagnosis from small samples in days is a key requirement. The evidence to support ‘liquid biopsy’ of cancers using circulating cells, nucleic acids and proteins for diagnosis, and to monitor the efficacy of treatment is fast becoming a reality.3
The effect of personalised medicine on NHS costs is still a matter of debate, but there is no question that for advanced cancer, most individualised treatment will add to costs. Patients will live longer, but sadly most will eventually succumb to their cancer, and in many cases this will be after multiple rounds of treatment, each one of which prolongs their life at a cost to the NHS. Such developments therefore need to be introduced in concert with strategies to improve early diagnosis of cancer and other disorders, when the chance of cure is high and less expensive.
The impact of improved scientific understanding of disease on pathologists practising in the NHS, university or commercial sector is daunting. Whilst the population has increased substantially, the exponential increase in the complexity of care is matched by increasing demands on diagnostics, which are expected to provide the scientific evidence necessary to optimise treatment. Medical imaging is providing its share, but the investigation of most molecules and cells requires a laboratory.
Changes already in progress
So how will departments of pathology change over the next decade to cope with increasing requirements in the face of decreasing or static budgets? I believe that some of the answers can be found in changes that started decades ago, including:
- clinical translation
- computing and information technology
- digital pathology and image analysis.
Automation revolutionised the way in which blood sciences labs work, leading to the amalgamation of clinical biochemistry and haematology laboratories. It of course makes sense that other blood tests from both immunology and microbiology are done in blood sciences labs, and this is increasingly the case. These highly automated laboratories are now largely staffed by scientists, with relatively few medical staff, most of whom now have greater clinical than laboratory commitments. The equipment is largely maintained by engineers from the manufacturers or third-party providers. One new development is that equipment footprints are shrinking, and as the size of instruments declines, power requirements are also lower. This has implications for laboratory size, throughput and placement.
Some specialties have reacted to this by becoming more clinical in their outlook. This now seems to be happening in genetics, which is splitting along clinical and laboratory lines, in the same way that haematology and clinical biochemistry did some time ago. It is possible that this trend will continue to involve other specialties within pathology. For instance, as the complexity of molecular diagnosis becomes greater, it is possible that doctors who started off in histopathology labs will find themselves in clinics, and making even greater contributions to multidisciplinary teams.
Automated laboratories are totally dependent on their laboratory information management system (LIMS), and most pathology departments now employ computing staff to ensure that this works optimally. This is one area that seems ripe for improvement. Most of our LIMS are databases trying to do multiple different jobs, and usually doing few of them well. As they increasingly have to interface with electronic patient records, as well as ever more complex equipment, the old database designs underlying many of them are showing the strain, particularly when asked to produce integrated reports,1 something that the College is actively exploring.
Immunohistochemistry can argue that it is the forerunner of modern molecular diagnosis. You don’t have to be that old to have seen the revolution in diagnostic accuracy and disease investigation that this technique brought to histopathology; it is quite clear that this is happening again with molecular methods, which are now commonplace in our laboratories. Immunohistochemistry of course grew out of histochemistry, which is perhaps set to enjoy a renaissance as mass spectroscopy enters a new phase of development.
Digital imaging systems are finally replacing microscopes in histopathology, and this is opening up the possibility that image analysis will enter routine clinical practice for the measurement of tumour size, grade and protein expression. This is the end result of advances in the manufacture of charge-coupled device (CCD) chips, and increasing computing power. These have found application in other branches of pathology, from plate-reading in microbiology to digital droplet PCR and next generation sequencing.
All of these changes have one thing in common: technological advance. This shows no sign of slowing and is probably accelerating. The face of pathology departments is changing, but it is interesting to note that despite the technological advances listed, no specialties have yet disappeared. Biochemists, haematologists, immunologists and serologists work side by side in blood sciences laboratories without losing their identities. In future, pathologists may be known by clinical colleagues more for their interpretation and reporting of results than their work in the laboratory. This is an important point. While it is impossible to hold back the tide, and most of us would not wish to, the expertise that resides with pathologists will always be needed, whatever their specialty, and overwork is much more likely to be the outcome of technological advance than underemployment!
Nowhere is this seen more clearly than with next-generation sequencing, which is in itself a hostage to fortune, as it is uncertain what the next generation of this will be called. In fact several very different technologies are already included under this label. However, the technique produces large amounts of data, which is a challenge to interpret and report in a manner that busy clinicians can cope with.
New technologies – what’s coming next?
Sequencing technology is going in several directions, and some form of sequencing technology is likely to be needed by all pathology specialties. Whole genome sequencing (WGS), perhaps at birth, may one day replace neonatal screening, provide lifelong pharmacogenetic information and allow risk-based disease screening. This may be done in large, automated laboratories, but we are already seeing the development of highly capable bench-top or smaller instruments that can provide rapid diagnostic sequencing of known mutations or microbial resistance genes. Automation of library preparation and reagent loading has reduced hands-on time to 45 minutes for one system already on the market, and another machine even plugs into a USB port on a computer, suggesting smartphone DNA diagnostics may not be far away?
Mass spectroscopy (MS) is now commonplace in microbiology, and blood sciences labs. Newer high-throughput methods and machines designed for specific purposes are increasingly capable of transforming diagnosis. MS has the potential to produce relatively inexpensive results, though the equipment currently requires major investment.
The ‘big data’ approach to data analysis and interpretation is already with us. Facilities with petabyte or even exabyte levels of storage are available, and can be accessed over networks. While these are currently mainly research facilities, it seems likely that they will eventually hold patient data, including whole genome sequences for use with outputs from local pathology labs or even near-patient (point of care) diagnostic devices. The idea of ‘linked-up’ pathology with digital diagnostic assistants and bioinformatics support from a distance is likely. It is certainly true that LIMS and electronic patients records (EPR) currently leave much to be desired, but the gaps in their provision are increasingly known and are likely to be filled. We need staff with bio-informatics expertise, as well as improved and updated computing facilities. Data security will continue to be an issue, but we now trust banks (and even social media, but perhaps not mobile phone providers) with our personal information, and it seems certain that these constraints will be overcome.
The near-patient testing landscape seems to be splitting into self-testing (e.g. glucose meters) and physician-use only machines, which might even in the end resemble a Star Trek-style tricorder, a handheld device capable of analysing multiple parameters at once. There is even a prize for anyone who can develop such a portable, wireless device that can be held in the palm of your hand that monitors and diagnoses your health conditions (http://tricorder.xprize.org). Whether the pathologist will appear on a small screen on the latter telling the physician how to use it, or which antibiotic to prescribe, is yet to be determined! Nanotechnology and implant technology are likely to combine with such machines to produce continuous, or at least easier, patient monitoring.
Clinical advances that will affect pathology
It never ceases to amaze that we have to take our cars in for an MOT and service every year (at our own expense) but that – as much more complex biological machines – people are not screened regularly for disorders, the treatment of which will cost the taxpayer much more. This is partly because there is insufficient evidence to justify general screening, and although the new ‘health checks’ in the 40–74 age group may go some way towards answering the need, screening must be a key area for future clinical development and will become increasingly sophisticated. I suspect that it will be cost effective and will perhaps help us overcome the cost implications of increasingly personalised medicine when we do fall ill.
Pathologists of the future may provide and interpret the screening tests, check the pharmacogenetic data based on WGS, monitor patients’ implants (remotely via their mobile phones) and request samples to ensure that treatment of infection or cancer can avoid resistance. This may sound like science fiction, and some of it may be, but it is worth reminding ourselves that the Cray supercomputer of the 1980s has the same performance as the Apple iPad into which I’m dictating this article right now!
Challenges for the future
Pathology is changing and, if anything, the pace of change is increasing. Changes in practice need to be closely aligned with a health economic assessment of their impact on patient pathways. Pathologists need to be increasingly innovative in developing their practice and have to lead this. Only by making the most of the opportunities offered by technology will pathologists and scientists in all specialties remain relevant, engaged and employed.
The opinions expressed here are those of the author, who hopes at least to generate a little discussion….
I am very grateful to all those who have contributed to this article, particularly Suzy Lishman, the College President, who found time to go though the first draft, and all those on the ISCMP and other committees within and outside the College who have helped formulate the ideas.
- Cree IA, Deans Z, Ligtenberg MJ, Normanno N, Edsjö A, Rouleau E, Solé F et al; European Society of Pathology Task Force on Quality Assurance in Molecular Pathology; Royal College of Pathologists. Guidance for laboratories performing molecular pathology for cancer patients. J Clin Pathol 2014;67):923-31. doi: 10.1136/jclinpath-2014-202404. Epub 2014 Jul 10. Review.
- Normanno N, Cree IA. Genomics driven-oncology: challenges and perspectives. BMC Cancer 2015;23;15:141. doi: 10.1186/s12885-015-1147-7.
- Cree IA. Liquid biopsy for cancer patients: Principles and practice. Pathogenesis 2015;2:1–4.