Q&A: Sentara's Tabetha Sundin on the Challenges of Molecular Technology in the Clinic

NEW YORK (GenomeWeb) – Last week, Virginia-based Sentara Consolidated Laboratory Services was designated a Roche Molecular Center of Excellence, one of 33 facilities to receive such designation, giving it access to molecular technologies from Roche that it may otherwise not have, as well as a voice to Roche as it further develops its technologies.

Sentara Healthcare, which comprises 12 hospitals, as well as urgent care centers, physician offices, and other healthcare facilities, first began moving into the molecular diagnostics space a decade ago. Yet, molecular-based testing still represents only a small fraction of the total number of tests performed by Sentara.

Cost considerations — both the price of the instruments, as well as for training personnel to use them — have posed barriers to adoption of molecular technologies. Additionally, despite advances in accuracy and speed offered by the new methods, older chemistry- and immune-based tests may actually have greater clinical utility for some circumstances.

Sentara Consolidated Laboratory Services Acting Director Tabetha Sundin spoke with 360Dx recently and said that while the use of molecular diagnostics is on the rise in the clinical setting, "the general patient walking in the door is probably not going to have a genetic test."

Below is a transcript of the conversation, which has been edited for clarity.


What is Sentara's role as a Roche Molecular Center of Excellence?

This gives us access to a network of other large molecular labs that have been named molecular centers of excellence with Roche. It allows us to interact with them, to talk about very specific challenges and new emerging technologies in our field and really collaborate across a large network with other large labs.

It also allows us to become early adopters of Roche technology, so if they have something new coming out, we could be a clinical trial site for them, or an early adopter, and that allows us to get access to the technology very early, provide input, and really mold where they go with that technology. And at the end of the day, if that gets FDA-approved … we would already have our validation complete. There's definitely a benefit to both of us.

Are there specific areas in which you will be working with Roche? For example, disease areas, applications, etc?

In our lab, we do a good mix of molecular pathology and molecular microbiology, so we have the ability to interface with them across the spectrum of diseases. We are new in adopting next-generation sequencing, and I believe Roche is going in that direction, so we hope to have the opportunity to work with them around sequencing. They do a lot with clinical virology, and we currently are already Roche users for clinical molecular virology, and we would look to provide insight on them, as well.

We do discuss emerging topics in our field with them and their networks, such as C. difficile testing algorithms, and we talked very vaguely about the backend of their sequencing data analysis pipeline that they are working on.

Do you have any sequencers at your facility?

We are new to next-generation sequencing, but we've been doing Sanger sequencing for quite some time. In our lab, we have a 3130 capillary electrophoresis [instrument by Sciex], we have a 3730 capillary electrophoresis Sanger sequencing analyzer, and we have recently adopted an Ion Torrent S5 and Ion Chef [by Thermo Fisher Scientific].

How are you using the Ion Torrent instrument?

We are currently live with an assay to analyze the CFTR gene for cystic fibrosis. We've been doing that since January, and we are in the end stages of validating the cancer hotspot panel available from Ion Torrent. We will go live [with the panel] in June.

What cancers will this be for?

This is all solid tumors.

What other molecular technologies does your lab have?

We do about 200,000 molecular tests per year, so we are pretty big for a molecular lab, and we do a broad array of molecular pathology and molecular microbiology. We use Roche to do EGFR, KRAS, and BRAF mutational analysis by PCR. We do quite a bit of molecular [microbiology]. We do chlamydia, gonorrhea, trichomoniasis, HPV, HBV, yeast, herpes. We do all the major viruses. We viral load testing for HIV, HBV, CMV.

We do some genetic testing with factor V, factor II and MTHFR. We also do T- and B-cell fragment analysis, JAK2 testing for AML. Fragile X testing we just went live with, also prenatal screening, so we do cystic fibrosis now and fragile X for prenatal screening.

We do HCV genotyping and HIV genotyping.

Aside from Roche, are there molecular platforms from other vendors that you use?

We use a great deal of Thermo Fisher with Ion Torrent and the [Sciex] genetic analyzers. For chlamydia, gonorrhea, trichomoniasis, we use Hologic. We've got two [of Hologic's] Panthers here. We use GenMark for genetic testing, factor V, factor II, MTHFR, and we also do HCV genotyping with them.

We have a respiratory virus panel that we use [from] Nanosphere. We use the Luminex for a smaller cystic fibrosis panel.

Is your molecular lab used only for in-house testing, or do you do testing for clients outside of Sentara?

We definitely have a large outreach program. We have a 12-hospital system, we have five freestanding emergency departments, and over 100 physician offices that are Sentara owned. We get a wide variety of samples.

We also own our third-party payor, Optima Health, so patients with that insurance, we get those samples, as well. We have one molecular lab for the system, centralized. Samples come to us from across our health system and from outreach, and that's what drives our high volume. We do quite a bit of esoteric testing here. If the test is even more esoteric, or a low volume, we do send that out to a reference laboratory.

When you assess new technologies, how do you decide which ones you may want to bring into the lab?

We really want to make sure that it's accurate, that there's good scientific evidence that it's a really well-founded assay, that there's good clinical utility. We look for a high enough volume that it makes sense to bring it in house. So, we are looking at normally our reference lab send-out list to determine what we want to bring in house. We look to see if we can do it cheaper than what we're paying to send it out.

We look for a great workflow. If we already have the instrument in house, it's a really easy decision to add testing to it.

[We ask if] it really makes good use of that space in our lab, and if my staff would already be competent on that technology. If it is new technology, we have to evaluate the space, the heat, the electricity, all the facility components to bring that in.

Molecular diagnostics in the clinical laboratory is growing at a very fast rate, so things that maybe were done previously in other departments — chemistry or microbiology or serology — they're moving a large volume toward molecular testing because of the rapid turnaround time that we're able to achieve. Patients can be discharged quicker or treated appropriately in a more timely manner. Maybe antibiotics [are] de-escalated when appropriate. We're seeing a huge increase in volume across the board.

There are new guidelines coming out in molecular microbiology and molecular pathology that drive molecular testing, especially in oncology, around the new companion diagnostics and complementary diagnostics to drive treatment decisions. All of those things really do increase our volume of molecular tests.

Would you say that molecular testing is part of the regular patient experience now at Sentara?

I think we are still pretty specific. In the oncology world, I would say absolutely. Every oncology patient is probably having some sort of genetic test, and [we're] moving more and more in that direction. Infectious disease — the patients are probably having some sort of molecular testing to help drive their treatment decisions, [whether] it's long-term management or diagnosis.

I would still say the general patient walking in the door is probably not going to have a genetic test, but I see that evolving over the next few years.

As you migrate some of your testing to molecular from chemical or immune-based, how are you making that decision?

Cost is a major factor, not just for us, but really for the patient. Patients are very aware of what their healthcare costs them, and we really try to make sure we are diligent in providing them the right test at the right time. In doing so, we want to make sure that every patient that walks in the door is not going to get a molecular test. It's the ICU patient with the respiratory distress that may need a molecular-based test or respiratory virus panel versus a flu swab.

For our C. diff algorithm with our hospital, we have really chosen to do kind of [a] triage. We will do a toxin and GDH, a low-cost test that really gives us two results. [If] they're concordant and both those results are either positive or negative, the patient is treated according to that result.

If they're discordant with one another, then it escalates to molecular testing. That way, the patients get the benefit of the molecular test on the back end if they need it. But we're not applying it widespread across our health system to make sure we're not driving up the cost of patient care.

Is there any way to quantify how much of your testing volume is molecular versus non-molecular?

Our reference lab does 9 million tests a year, and we do 200,000 molecular tests a year. It's still a very, very low percentage.

Do the variety of facilities within Sentara's network differ in their use of molecular technology?

In physician offices, they're using a lot more point-of-care testing, something that they have in their offices that's very easy to operate by the staff that they have, and they can get an answer back to the patient before the patient leaves. Molecular testing is moving into the point-of-care field, but I don't know that those are deployed across our health system yet. Right now, there's almost no molecular testing being done in a physician office, but we work with quite a few specialties — women's health, oncology, and infectious disease — that order a high volume of molecular testing, and that really drives their patient treatment.

Women's health is probably our second-biggest user of molecular technology, which doesn't seem intuitive. Oncology is very specific, they really need a molecular answer to be able to treat a patient correctly, but women's health, really STI testing … we do that at a really high volume. I would say more than 60 percent of our volume is women's health volume.

They're doing that by molecular methods and some of those bacteria are hard to culture, or it would take too long and cost too much, so they're ordering these large panels of tests that we're able to perform very quickly by molecular methods, and the patients in a matter of three days could have answers for all the tests on our panel and be treated appropriately. That way they're not misusing antibiotics.

Under what circumstances would molecular testing not be appropriate? 

The limitations for molecular, there are really two: it's cost prohibitive, and we can detect a dead bacteria or virus. A positive-by-molecular assay does not mean there's an active infection there. The infection could have cleared and we would still get a positive result. We really work closely with our microbiology specialist, Kathy Judge, to make sure that we've made an informed decision about when molecular would be the right test and when molecular would be overkill.

And that's why we get together and develop algorithms or adopt algorithms that have been developed out there in the literature to make sure that our patients are still getting the benefits of the molecular tests, but we're not driving up the costs of healthcare.

As you're working with some of these molecular technology developers, what are you telling them in terms of your needs?

In our molecular lab, space is always a consideration, so we're looking for smaller instruments. We are seeing a trend toward primary tube-to-answer to decrease tech time. Molecular techs are very expensive to train, hard to find, and we really want to increase automation in our lab, provide better patient care, and we want to drive down costs.

I think as we are able to get instruments that run a lot of assays on that same instrument with a great workflow and a cheaper price — we've seen the cost of molecular microbiology testing come down significantly — as molecular pathology, molecular testing becomes more prevalent, we would expect those costs to go down on a cost-per-test basis based on the increase on volume.

We are a reference lab housed in a hospital, so there's never enough space. The companies that are coming out with primary tube-to-answer assays that have great workflows with a large test menu on board, some of those instruments are still very large, so those are huge decisions. We would have to do a lot of construction in our laboratory in order to fit those instruments.

What about personnel issues? Do you have problems finding the right people for your lab?

We actually have grown in the past four years from three technologists to seven technologists in the molecular lab, and I have an additional seven in our serology lab, and they perform some of our molecular microbiology testing. We have a fairly small staff, but we have trained them in house. I believe that we were able to hire three technologist with previous molecular diagnostics experience. Everybody else, they were either med techs, or recent biology college graduates with Bachelor's degrees that we've been able to train for clinical molecular diagnostics testing.

What is your reimbursement experience?

We are in a Palmetto state, and we've had to apply recently new Z codes to molecular billing to be reimbursed. I think there's always been a struggle for molecular clinical labs to be reimbursed. Molecular microbiology is generally reimbursed well and molecular pathology is reimbursed poorly. We try to balance that. Molecular microbiology volume is high, cost is low, and reimbursement is good, so that's a good win for our lab. Molecular pathology is the reverse. It's expensive, it's low volume, and it's not reimbursed well. And so those two generally balance themselves out in our lab. We make sure that as we go live with a new molecular pathology test, we bring on new molecular microbiology testing around the same time to kind of balance them out.

For those in the clinical setting looking to bring molecular testing in house, what kind of advice do you have?

There are a couple of different things. There are a lot of manufacturers that are coming out with really moderately complex instruments that are sample to answer, which is very nice to decentralize in the molecular lab. Those things can go into a medium-sized microbiology lab and really improve patient care. I think that's a great thing.

I would say that if you have no experience with molecular testing, you have to be very careful about the different kinds of contamination, because contamination to a molecular biologist is very different than that to a microbiologist. In our lab, we have to have negative and positive air flows, we have separation of work space, and we really have the facility to make sure that if we have a highly complex molecular test where amplicons may be manipulated that we can do that.

I would also say be very careful and understand what your result means before applying molecular testing universally, to understand when a microbiology test should be used, when a molecular test should not be used, and when an algorithm should be employed to make sure that you're using molecular testing appropriately.