2014 Emerging Clinical & Laboratory Diagnostics: The Portable Lab

For decades, the clinical laboratory has been recognized as the leading model for applied analytical chemistry labs. Clinical chemists are the first to deal with real-world concerns such as large-scale automation, high throughput, and infinitesimal turnaround time, and resolving issues with information technology including data quality, data curation, analysis report generation, and dissemination. These are all current topics in chemistry laboratories, but most were addressed first in clinical labs. Plus, the clinical labs have blazed a trail through an inhospitable, regulated, and cost-conscious environment. This year, these topics and more attracted about 200 scientists to the Fairmont Hotel in San Jose, CA, from April 24 to 25.

The portable laboratory/point of care

The portable laboratory was the stated focus of this year’s meeting. Some of the drivers for the portable laboratory are: point of care (POC) analysis to improve turnaround time or to comply with the realities of resource-limited countries. Since half of the patient populations in the developed world use smartphones, new paradigms are in communication between physician, patient, payer, and public health. Others were attracted to POC technology, but with products that plug into the wall for electricity. These seem to be designed for physicians’ office laboratories or smaller clinics in remote locations.

Breath analysis

Despite utility dating back to Hippocrates, breath analysis as a matrix for diagnostics is less popular than blood and urine, and even saliva. A fascinating lecture from Michael Phillips, M.D., of Menssana Research (Fort Lee, NJ) traced the history of breath analysis from Hippocrates to date. Hippocrates had connected the dots between breath that smelled of rotten apples to diabetes, urine odor to kidney failure, and sewer smell to lung abscess. Fast forward to the 1700s, when Lavoisier proved that breath contained CO2 by bubbling breath through a solution of Ca++. Fast forward again to 1971, when Prof. Linus Pauling (California Institute of Technology, Pasadena) used a cold trap attached to a gas chromatograph to show the presence of more than 100 compounds in breath.

Analysis of human breath has been problematic primarily due to sampling. One needs breath from the lungs. This occurs later in the breathing cycle. Dr. Phillips has developed a long tube that latterly segregates the breath in time.

Initially, in the research phase, he analyzed breath with GC/MS. He became convinced that coelutions were a frequent problem, so he resorted to 2-D GC × GC/MS. Suspicions were confirmed: Up to five other analytes coeluted with hexane in the first dimension.

Since GC × GC/MS is too expensive and complex for POC assays, Dr. Phillips developed a small, 1-D GC using an array of surface acoustic wave detectors for the field-deployable analyzer. Since retention times in automated GC are very reproducible (Tr CVs of less than 0.1% are achievable) one can use the differential response of differentially derivatized surface acoustic wave (SAW) elements to determine the composition of a composite peak, such as hexane. He showed a figure of a current system, called BreathLink™ (patent pending, Menssana Research). The GCs and sampling tube are housed in a workstation. It reminded me of productization of a working breadboard.

The BreathLink was used to screen the breath of high-risk tuberculosis patients. The chromatograms showed that the positive predictive value was only 11.8%, but the negative predictive value (NPV) was 98.8%. Thus the probability of a false negative is very low. So the BreathLink is useful in screening a large population. Since the usual prevalence of TB is only a few percent, the large majority of the cohort can be confidently ruled out. Attention is focused on a much smaller cohort with positive but ambiguous results. A similar GC assay for volatile organic carbon (VOC) markers produced a NPV of 99.9%. This is comparable to the NPV of mammography but avoids the discomfort and radiation exposure. Plus the cost per group of patients is reduced. Currently, the BreathLink is being evaluated with larger cohorts to meet the FDA’s requirements.

I expect that the next-generation design will be much smaller and integrated into a single, human-portable device. In two or three generations, it will be a mask that fits over the nose. Data will be curated and transmitted to the giant processor in the cloud for near-real-time evaluation and report generation, which will return via the reverse route before the subject has time to wander off.

Paradigm shift enabled by smartphones

Developers see the smartphone as an interesting platform that provides a common, well-engineered, and familiar human interface. The specific implementation depends upon purpose and available infrastructure.

In the developed world, compliance with therapy is one of the weak points in the current healthcare paradigm. Physicians prescribe a therapy for a particular malady. About one-third of prescriptions are never filled. Even when the Rx is filled, many patients cease using the drug because it is not producing any improvement, or the problem is cured, i.e., “Symptoms gone so I don’t need it any more….” However, on the next visit to the doctor, the patient won’t admit to noncompliance.

With modern IT, one can be fitted with a sensor with two-way communication that provides the physician with frequent updates on macro indications such as body temperature, blood pressure, pO, etc. With properly designed assays, this could be extended to particular drugs or metabolites. A fascinating lecture by Joshua Windmiller, Ph.D. (Electrozyme LLC, La, Jolla, CA) discussed fabrication-wearable biosensors for a variety of analytes. Examples included tattoos of electrical sensors based upon electrochemical reactions that sense lactate, ammonia, sodium uric acid, and pH on the skin surface.

Suzanne Sysko Clough, M.D., Founder and Chief Medical Officer of WellDoc (Baltimore, MD), started her lecture with a report from Eric Topol, M.D., in which he says he is prescribing more apps than medications.1 He predicts that “the digital revolution will create better healthcare.”

How? Dr. Clough points out that the infrastructure is in place. By 2015 it will include more than one billion smartphones, one billion computers, and 400 million tablets. This will enable mHealth (mobile health) with empowered patients and providers. The drivers include cost reduction, improving: 1) engagement, especially of the patients; 2) satisfaction; and 3) outcomes. The infrastructure is indeed in place—most people check their smartphones several times per day.

WellDoc’s BlueStar Diabetes Mobile Prescription Therapy is an FDA-approved smartphone application that incorporates four core modules. The first is medication management including dose calculation. The second is symptom management, where the patient reports sweating, nausea, etc. Lifestyle management is next. This module helps the patient with coaching on diet, exercise, stress reduction, etc. The last is physiologic management including blood glucose, A1c, heart rate, and lab results. The smart communication helps build patient compliance. The managing physician has a near real-time view of patient status and compliance with a visual dashboard. Dr. Clough reports that patients and physicians see significant improvement in compliance and attitude.

FDA regulation of smartphone technology is an issue addressed by Scott Cunningham, J.D., of Covington Burling LLP (Washington, DC) and Dr. Clough. Since the state-of-the-art is dynamic and changing rapidly, and there are more pressing issues, the FDA is utilizing selective enforcement based upon risk-based considerations to focus on products with a high risk to patient safety. Approved items include MobiUS SP1 Ultrasound (Mobisante, Redmond, WA), Proteus Helius Pill Sensor (Proteus® Digital Health, Redwood City, CA), AirStrip OB (AirStrip®, San Antonio, TX), and AliveCor ECG (AliveCor™, San Francisco, CA). Dr. Clough also reported that BlueStar is FDA approved, but since they were one of the first, they worked with the Agency for almost a decade.

Resource-limited settings

WHO recognizes the need for a significant paradigm shift for clinical diagnostics to be compatible with the resource-limited world’s lack of infrastructure but huge for improved healthcare. This flies under the acronym of “ASSURED,” which stands for Affordable, Sensitive, Specific, User friendly, Rapid and Robust, Equipment free, and Dependable.

Novel handheld platforms for POC assays

This was addressed by Dr. Courtney Nicholson of AgPlus Diagnostics (Sharnbrook, Bedfordshire, U.K.) in her lecture on the company’s assay platform, which is designed to use saliva as a sample matrix for rapid, quantitative diagnostics in low-resource settings. The analyzer has demonstrated LODs of 15 pg/mL testosterone, 1 ng/ mL cortisol, and 20 virus particles.

Of all the new platforms that were discussed at the meeting, I was most impressed with the Multiplex POC platform for biomarker detection developed by OJBio Ltd. (a joint venture between Orla Protein Technologies and Japan Radio Company, Newcastle upon Tyne, U.K.).2 The instrument utilizes highly selective SAW technology for focused assays of biomarkers. In operation, the highly engineered microchip holder is placed in the handheld human interface. The liquid sample is applied to the well (blood, saliva, urine, etc.). The solution is wicked to the SAW pads. The signal is sent via Bluetooth to a smartphone or other transceiver for processing and report generation. The chip is then removed and discarded.

Both platforms above are communication enabled, i.e., cell phone compatible. Upgrading to smartphone technology is expected to be rapid, especially in population centers. Also, the low data rates of cell phones, as opposed to smartphones, favor processing at least some of the data at the point of care.

Several lectures and posters addressed the use of smartphones in general and iPhones in particular. Generally these involved POC situations. Dr. Onur Mudanyall of Holomic (Los Angeles, CA) reported that 7 billion cell phones are in use globally with 5 billion phones in developing countries. Some have more active cell phone subscriptions than people. Later, Dr. Toby C. Cornish of Johns Hopkins (Baltimore, MD) pointed out that cell phone coverage is better in Africa than he experiences in the U.S.A. So cell phones are ubiquitous, but smartphones are replacing dumb ones around the world.

Holomic has developed several assay platforms that use a smartphone as the human and communication interface. The simplest is a rapid test reader. Another platform runs assays for allergens, mercury, and urine. Still another offers lens-free holographic microscopy. The reader is used to capture images from lateral flow immunoassay tests with a detection limit of 0.005 OD. Assays include aflatoxins and HIV. Postrun processing is performed in the cloud.

Several posters from PATH (Seattle, WA) described a complete workflow for isothermal malaria loop-mediated isothermal amplification (LAMP) assay. The key is developing a heating source for the amplification that uses phasechange material coupled with an intense exothermic chemical reaction to support linear, isothermal amplification. The time required from fingerstick to report is less than 1 hr.

New traditional POC platforms

POC assays platforms need to meet the ASSURED criteria, but the particular execution may be different since reliable electricity and water are usually available and inexpensive. QIAGEN (Valencia, CA) introduced the ESEQuant LR3 for POC fluorescence readers, particularly for lateral flow assays. The ESEQuant TS2 is designed for fluorescence measurement of isothermal nucleic acid amplification assays.

TetraCore (Rockville, MD) introduced the T-COR8 for quantitative PCR (qPCR). TetraCore’s goal is to make qPCR sufficiently easy to use and reliable that the T-COR8 would qualify for a CLIA waiver.

Microring resonators are novel solid-state devices in which evanescent waves from a laser pass near the optical microring on a silicon chip. Microrings (30 μm diam with 200-nm-high ring) are fabricated using deep UV lithography with about 600 chips on an 8-in. wafer. In resonance, the photons circulate around the ring many times, giving a sampling length much longer than the structure. Ligands are bound to the ring surface. When an analyte molecule binds, it produces a shift in the resonance wavelength, which is easily measured. This provides label-free detection. Arrays of rings can improve detection sensitivity or extend the range of analytes (multiplexed assays). Detection limits can be improved by adding reporter molecules to the surface. The chips are low cost and disposable.

The Maverick platform from Genalyte (San Diego, CA) automatically processes samples. It uses fast scanning mirrors and lasers to interrogate chips that can have as many as 128 rings. The detection limit with label-free analytes is about 2 ng/mL; processing time is about 10 min. Enzyme-linked immunosorbent assay (ELISA) takes 3 hr. Adding secondary and tertiary binding events improves detection limit to about 30 pg/mL (~200 fM) but at the expense of increased processing time to about an hour. If even lower detection limit is required, enzymatic signal amplification can achieve LODs to less than 1 pg/mL, but analysis time is about 100 min. Assay cards for the Maverick process two samples independently. Each analyte channel uses an array field of four sensors per analyte with up to 16 analytes per sample channel.

Nonspecific adsorption

Nonspecific adsorption (NSA) is a problem in immunoassays of proteins in multiple-well plates. Proteins, be they antagonists or antibodies, can adsorb on the bottom of the plate in various forms. This adsorption is not uniform from well to well. A lecture by Robert Matson of QuantiScientifics, LLC (Irvine, CA) pointed out that proteins on a long tether behave more like free-solution proteins. He constructs the tether by first printing and covalently bonding a segment of single-stranded DNA to the bottom of the well. If this DNA has a site-specific sequence, then one can attach the compliment to this site. If the compliment contains the agonist or antibody, then it will be localized at this site also. Multiplexing is easy by microprinting on the well bottom. A 96-well plate with 13 spots/well gives 1248 assays.

Genome sequencing

The last lecture of the meeting was a kaleidoscope on the status of genome sequencing for research by Stephen Turner, Ph.D., of Pacific Biosciences (Menlo Park, CA). The company introduced a long read sequencer in 2010 using sequencing by synthesis utilizing light emitted by the reaction in the ribosome. The key advantages are long read length, calling accuracy, and speed.

With over 100 instruments in the field as of the end of 2013, the most useful advantage is avoiding and indeed correcting alignment and assembly errors that occur with shotgun technology. Read lengths are as long as 20 kb in 2014, which is long enough to read through long segments of repeats. In one case they found re-sorting of a gene from another host, which explained the food poisoning in Germany a few years ago.3

Looking to the future, Dr. Turner predicted that in the next few years their read length Emerging Clinical & Laboratory Diagnostics will exceed 30 kb, and soon will be capable of whole chromosome reads, including even the centromeres. Today, centromeres are not routinely sequenced, but indications are that they show high variability compared to the long arms of the chromosome.

DNA methylation is a controversial topic. Dr. Turner showed that the read methylation of DNA decreases the read speed. The exact effect depends upon the particular methylated isomer in the DNA. What this means is not clear, but the ability to read through and classify DNA methylation should be useful for understanding the biochemistry.

While Dr. Turner was clear that his lecture was a view of the status of research, most understood that the research would impact clinical diagnostics in the near future.

Conclusion

At the end of the meeting, a clear picture came into focus. The functions of the each of us can be monitored perhaps even continuously with tattoos, wearable undergarments, and breath analysis. Should one feel ill, or just visiting the physician for a checkup, more data-rich point of care screens using GC SAW detection combined with saliva monitors will provide quick, data-rich analytics. The data will be curated on site until ready for transmission by microbursts to satellites, smartphones, the mesh, and onto the cloud for processing. Postprocessed reports are transmitted back to the source within a few minutes. Depending on the scenario, the patient can be told of the result and treated appropriately.

Credits

For more than four decades, the American Association for Clinical Chemistry (AACC) organized the Oak Ridge Conference on the future of technology relevant to clinical chemistry. Over time, the “Oak Ridge” part of the title became less relevant. So this year, the meeting was retitled “Emerging Clinical & Laboratory Diagnostics: The Portable Lab.” The mission statement for the meeting is probably very similar. Attendees have consistently rated this series as the best place to talk shop on technical developments in clinical diagnostics.

References

  1. Topol, E. The Creative Destruction of Medicine, How the Digital Revolution Will Create Better Health Care; Basic Books, New York, NY: 2013. ISBN: 978-0465061839.
  2. http://www.oj-bio.com/
  3. http://en.wikipedia.org/wiki/2011_Germany_E._coli_O104:H4_outbreak

Robert L. Stevenson, Ph.D., is Editor, American Laboratory/Labcompare; e-mail: [email protected].