The disease discovery and therapy workflow is lengthy and convoluted, with the potential for technical, regulatory and turf issues. Shane Needham, Ph.D. (co-founder and laboratory director of Alturas Analytics, Inc. [Moscow, Idaho], a bioanalytical CRO focused on LC-MS/MS and GC-MS/MS), maintains that once a patient has been diagnosed with a chronic disease, the plan shifts to therapeutic maintenance. In a lecture on “Analytical Technology: From the Analytical Chemistry Bench Straight Results to Patients: Is This Really True?” Shane centers on patient need, then searches for the most expedient way to meet it. His proposed model would reduce the delay time between discovery of a reported diagnostically significant correlation of symptom with a causative marker and utilization of this relationship to improve patient outcome.
RLS: How did you become interested in improving the workflow from the chemistry lab to the patient?
SN: The profitability and competitiveness of our lab depends upon operating efficiently. Over the decades, I’ve developed a sense for bottlenecks and keyholes from incoming receipt of samples to delivery of actionable information to the work site, patient, etc.
However, my interest really got personal when my son was diagnosed with type 1 diabetes. The initial regimen was routine, many-times-per-day glucose testing, followed by quarterly doctor visits and adjustments of therapy. However, if my son really wanted to beat this disease, he would need to have the tools to adjust his own therapy continuously. Thus, we started to investigate.
RLS: What did you find?
SN: We started [reading] study literature and keeping journals of his glucose levels and how they correlated to how he felt, his athletic performance [and] his ability to think quickly. Like scientists, we kept records of all of this. For optimal performance, we knew his levels must be between 80 and 120 mg/dL. These are typical levels of a healthy patient but difficult to achieve for a type 1 diabetic. We knew it was possible.
We started to actively manage his glucose level, including being meticulous about what he ate, his daily cycles, influence of stress hormones, his insulin regime, his exercise regime, etc. Soon, good days were more frequent. The normal range for a large cohort of individuals with diabetes is 80 to 120 mg/dL. This is fine, but the effective therapy range for some individuals is much narrower, and, apparently, at least possibly, individual specific. I hypothesized that there is an advantage to individualized therapy.
After watching my son improve his health with continuous monitoring of the biomarker, glucose, I knew other therapies could be personalized to improve treatment if, as analytical chemists, we give patients the right tools to monitor therapy and self-treat. This truly is personalized medicine with point-of-care diagnostics and treatment.
This point of care use wouldn’t be possible without a personal glucose monitor. I want us to now develop other devices to measure more biomarkers for point-of-care use and treatment.
What I also found was that, after much research and talking with medical providers from the 1950s and 1960s, the medical community used to say things like, “[P]atients shouldn’t be able to check their own blood pressure or monitor their own blood glucose due to not knowing proper treatment.” Look at how far we have come and how many lives have been saved by self-monitoring and self-treatment of these two markers.
RLS: Individualized therapy with a narrow therapeutic window seems unusual given the wide range of therapeutic responses published in the literature. Any comments?
SN: You raise a very good point. Let’s look at some specific areas of personalized medicine. Take the treatment of bipolar depression with lithium salts. The therapeutic range for efficacy of lithium in the blood is 0.8 to 1.1 mmol/dL. Below this range, lithium has little effect. Above this range, major side effects can occur.
The sex hormones testosterone and estrogen also have narrow windows of efficacy that may be individual specific. This requires a dose response determination for each individual. The very broad “normal” range of testosterone really reflects the variability in the response of a population. For example, age is a factor, so one looks for the age-adjusted values to guide therapy. Ranges of these hormones also change throughout the day, so continual personalized measurement could be a huge advantage to optimal health.
The desired endpoint is another factor. Natural production of sex hormones declines with age. So one potential question is: Is the decline in a particular patient abnormal? Does the decline fit the symptoms? Does the decline adversely affect quality of life? If so, then one can consider supplementing normal production with transdermal patches or IM [intramuscular] injections. This can improve the quality of life.
Imagine the quality of life that would be possible if we had a handheld device the size of a smartphone that could monitor levels of these biomarkers or therapeutics and we can adjust our therapy in real-time, personally. This truly is personalized medicine!
RLS: What kind of instrumentation do you foresee for rapid and reliable assays?
SN: We need a separation and detection technique. Capillary electrophoresis [CE] seems to be most suitable separation technique since one can create capillaries on a chip. This avoids the pressure that is a consequence of HPLC. Lab-on-a-chip assays are attractive for speed, reliability and small size. CE on a chip offers low solvent consumption and attractive detection sensitivity due to the solvent flow rate.
I’m also optimistic about ion mobility spectrometry (IMS). IMS is fast and [is] performed at atmospheric pressure; thus no large, costly vacuum pumps are needed. I think in order to introduce ions into the IMS, nanospray [electrospray ionization] (nESI) will be the technique of choice.
The future handheld device for personalized medicine could likely be CE-nESI-IMS.
RLS: Why IMS? Wouldn’t one prefer conventional MS?
SN: Conventional MS involves a vacuum. Vacuums require pumps, which require lots of space, lots of electrical power, and are expensive. IMS can focus on a few target analytes, as we see in airport scanners.
Diagnostics usually involve specific targets, so the versatility of a “large” research mass spectrometer is not necessary when going after targeted assays.
RLS: How do you see the work and information flow for assays specific to an individual?
SN: For less targeted, “nonpersonalized assays,” where the original diagnosis occurs before maintenance treatment, I see that the lab would receive a sample and work request, as is done now. The appropriate assay and instrument would be selected depending upon the unique needs of the patient.
General-purpose instruments such as IMS, LC/MS/MS or even GC/MS/ MS would analyze the sample. A report including the diagnosis would be prepared based upon the data and patient history. The report would be transmitted simultaneously to the patient’s cell phone. If desired, the patient could also have the data sent to the physician [to generate] electronic medical records.
RLS: This sounds good, but how does this differ from current work and information flow?
SN: In the developed world, particularly in Europe, the proposed work and information flow would not involve large changes, since many research labs also have an applied clinical function. However, in contrast to many automated clinical analyzers, mass spectrometers can be programmed to perform several different assays by simply changing the MRM [multiple reaction monitoring] to focus a targeted analyte, biomarker or therapeutic. The MS could use automated liquid handlers to perform a common sample prep protocol such as spin, dilute and shoot. The MS and data station would take care of the rest. The key for the future will always be focused on patient-driven healthcare. As analytical chemists, we need to help provide patients with the tools to drive their health decisions with their physician and other medical providers.
RLS: How would this same concept work for unregulated therapy such as fecal transplants?
SN: Fecal transplants are an interesting case. Antibiotics can disrupt the microbiome. Reestablishment of a microbiome may require searching the microbiomes of several candidate donors. Following antibiotic treatment, the intestine of the patient can be reinoculated with a sample of the microbiome of a closely matched specimen. What is most interesting about this story and the easiest way to explain this is that, although “unregulated,” the regulators have seen the life-saving results provided by fecal transplants, so these treatments are approved by default. This is important because as we provide solutions to personalize healthcare, and provide subsequent data to support this evidence, these devices and solutions will be fast-tracked for approval. Consumer-driven healthcare will demand these products.
RLS: Your vision seems to focus on developing devices for point-of-care use and treatment with the first diagnostic from a core facility in a central hub. In contrast, others are distributing low-cost clinical assays to the consumer for testing. Is this right? In what scenario does your model fit best?
SN: I’m impressed with the innovative designs using paper and printed reagents. These are probably suitable for basic tests, such as HIV, pregnancy, liver function, etc., in rural settings. These assays will probably be used to rule out patients in screening scenarios but really can’t be used for precise treatment since the results are so subjectively analyzed.
In third-world sites where they can have large populations, a central laboratory may make sense as a first step in building a modern science-based healthcare system or for diagnostics before maintenance treatment.
Versatile instruments such as MS/MS, HPCE/MS and multiplexed quantitative PCR [polymerase chain reaction] can provide rapid, high-quality analytics for [an] urban populace.
Rural patients can be directed to an urban hub laboratory. One key success factor is to provide results during the same visit. A turnaround time involving overnight or longer will probably decrease compliance.
Just [as] in any economy, consumers will drive this decision, so let’s give them all the options and let them decide what works best. Let’s give them the option for personalized point-of-care use and treatment with one device for many treatments. I think this leads to the best outcomes.
However, if consumers want to collect and send their sample to a core facility and it is economically viable, let them choose this option. There may also be the simple tests (HIV, pregnancy, etc.) that are still available as one-off tests in the local pharmacy.
RLS: The popular press reports using printed paper tests and 3-D printers to rapidly expand coverage and reduce cost.
SN: Yes, these are intriguing technologies and may fit in specific situations. However, I’ve not been impressed with “dip and read test papers,” except for qualitative analysis. The dynamic range of paper seems limited. Plus, if one is utilizing the human eye for the signal transducer, the interindividual variance is too large for quality data.
3-D printing for chip-based assays sounds attractive until one looks at the complexity of manufacturing the chips. Special care is required to control nonspecific adsorption of the analyte. Controlling electroosmotic flow is often a devilish problem. However, let’s continue to research this area and see what develops.
RLS: Earlier, you mentioned mass spectrometers and IMS instruments. Are the instruments robust enough for daily use? What about the required skills of the operator and report preparer?
SN: Mass spectrometers and ion mobility spectrometers are robust enough for routine use, as shown by airport scanners, food inspectors, et al., that already use these instruments.
The skills required to use the CE-nESI-IMS instrument that I envision would be no more difficult to operate than a cell phone that doubles as a blood glucose monitor (and these instruments exist).
RLS: You said that the workflow could be improved by directly sending data from chemistry labs (personalized device or core facility) to the point of care. How would this save time?
SN: Chemistry labs are often well-equipped with versatile instruments such as MS. The MS can use high resolution and MRM to provide quantitative assays of many analytes. There is no need for intermediate steps and products that require FDA approval and monitoring. To a limited extent, this model is being used in situations outside the U.S.A. In the U.S.A., the FDA’s proposed laboratory developed test protocol may hinder this type of operation. We need to work together to ensure that the patient receives data as quickly as possible. The patient[s] then can drive the decision and discuss this with their healthcare provider instead of waiting for days or weeks for the provider to contact them.
It is important for labs to think about the amount and kind of regulation that makes sense for the patient. The patient is the customer.
RLS: Do you have anything else to add?
SN: The future of healthcare will be driven by consumers, i.e., patients. As analytical chemists, we should embrace this and develop the best tools to optimize patient outcomes.
Robert L. Stevenson, Ph.D., is editor emeritus, American Laboratory/ Labcompare; e-mail: [email protected]