Molecular Diagnostics: Growing Pains in a Maturing Field

By Renee DiIulio

Molecular diagnostics laboratories are no longer the sole purview of large academic and research institutions, but as they are incorporated into smaller facilities, experts consider clinical utility, volume, staffing, and expense.

Molecular diagnostics emerged in the 1990s and is still a maturing field. Referring to the tests and techniques now available, Daniel H Farkas, PhD, HCLD, director of molecular diagnostics, The Methodist Hospital (Houston), and associate professor of pathology and laboratory medicine at the Weill Medical College of Cornell University (New York), says, “We’ve picked the low-hanging fruit. The amplification of these technologies into other diseases is slow.”

“For many administrators, molecular diagnostics is still pie-in-the-sky stuff done at NIH [National Institutes of Health in Bethesda, MD] and big centers, but it can be done at hospitals too,” says Jan A Nowak, MD, PhD, medical director of the molecular pathology laboratory of ACL Laboratories (West Allis, Wis), and a pathologist with the Advocate Lutheran General Hospital (Park Ridge, Ill).

Pushed by the expansion of applications and commercially available products, molecular diagnostics labs are slowly making their way into community hospitals and other smaller facilities. But this growth is limited by the usual factors, such as budget and staffing shortages, as well as some field-specific factors, including intellectual-property rights, physician education, and an uncertainty regarding some of the tests and their clinical utility.

Jeffrey A Kant, MD, PhD, FCAP, FAAAS, University of Pittsburgh Medical Center

Despite the challenges, or perhaps because of them, the molecular diagnostics lab is ideally positioned to work with all members of the clinical team, including other labs and physicians. “Molecular diagnostics is a great place for partnerships,” says Jeffrey A Kant, MD, PhD, FCAP, FAAAS, professor of pathology and human genetics, and director, division of molecular diagnostics, at the University of Pittsburgh Medical Center.

Personalizing Medicine
“Historically, molecular diagnostics dealt with infectious-disease detection,” says Nowak, noting that early tests targeted diseases such as HIV that could not be detected in other ways. Eventually, tests were developed for conditions, such as group B streptococcus, that were already detectable.

But molecular diagnostics is not only useful for detection—it can also be used to predict disease and determine treatment. Genetic testing can be used to identify gene-sequence abnormalities that may result in disease, such as BRCA1 assays that identify women at risk of developing breast cancer. Pharmacogenomics focuses on targeting enzyme sequences that can help to determine whether a drug will be effective and at what dose.

“We all have DNA-sequence variations, including in areas that code for the enzymes that metabolize drugs. If a patient has a variant that diminishes a drug or speeds it up, you need to adjust the dose to obtain the therapeutic effect,” says Kant.

The ability to predict disease likelihood and drug interactions contributes to the idea of personalized medicine but also raises privacy and responsibility issues. While the ethics are debated, the technology to achieve these capabilities continues to advance. And while some tests are considered too esoteric for commercial development, high-volume tests, some featuring automated equipment, are now offered by manufacturers.

Roche’s COBAS Amplicor automates PCR by combining five instruments: the thermal cycler, automatic pipettor, incubator, washer, and reader.

One such product is the COBAS Amplicor from Roche Molecular Diagnostics. It automates polymerase chain reaction (PCR), combining five instruments: the thermal cycler, automatic pipettor, incubator, washer, and reader. The machine can run 48 samples at once and return results within 1 day for nine diagnoses: Human Immunodeficiency Virus Type 1 (HIV-1), Hepatitis C virus, Hepatitis B virus (HBV); Cytomegalovirus, Chlamydia trachomatis, Neisseria gonorrhoeae, Mycobacterium tuberculosis, Mycobacterium avium, and Mycobacterium intracellulare. The turnaround is faster than conventional methods, and the results show greater reproducibility. The COBAS AmpliPrep automates the preparation of samples for the COBAS Amplicor.

Roche’s COBAS AmpliPrep replaces manual sample-preparation steps for PCR.

Other manufacturers offer similar products, and there are items in the pipeline, including newer technologies such as chips and microarrays. “The continuum is moving from manual to automated, principally because the market is being addressed by manufacturers. There are many nucleic acid amplification tests, including PCR. Currently, we have first- and second-generation tests, but the third and fourth generations are coming down the pike. These tests are allowing more hospitals to bring molecular diagnostics testing in-house,” says Farkas.

Jim Floberg, Roche Molecular Diagnostics

“Efforts to improve automation are intended to make molecular diagnostics more accessible and advanced,” says Jim Floberg, director of product marketing, molecular diagnostics, with Roche.

Improving With Age
With faster turnaround and greater reproducibility come improved patient care and less expense, causing hospitals to consider bringing at least some of this testing in-house. “Some of the newer instruments are inexpensive and straightforward enough that with high-volume tests, there can be significant cost savings,” says Kant.

“Molecular diagnostics is now very much a part of clinical lab services. Ten years ago, assays were complex and could take a week or longer to produce results. Now, it’s much shorter. ACL performs 1,500 tests a month, and the longest turnaround time is 2 days,” says Nowak.

He notes that stat testing is now a possibility. “Faster turnarounds mean hospitals can conduct successful screening programs, such as those for methicillin-resistant Staphylococcus aureus, to combat nosocomial infections and drug-resistance concerns,” says Nowak.

Kant reports an annual 20% to 25% growth in the samples processed by his lab without a significant increase in the testing menu. “We’ve been teaching this field to pathology trainees, and the technologies have been exposed at meetings, so physicians are more comfortable with the tests,” says Kant.

The Learning Process
This comfort, however, doesn’t necessarily translate to knowledge. Molecular diagnostics labs find themselves in the position of having to educate physicians about both the tests that are available and how to interpret the results. “Labs need to make sure the customers they serve can understand the rich data coming from these tests,” says Floberg.

“Clinicians don’t know the limitations or advantages of the tests, and they need that level of professional management and clinical interaction. They may not even know that certain tests are available in their labs,” says Nowak, who advocates more interaction between physicians and the lab.

If physicians are aware the tests exist, they may order them unnecessarily. “There is a tendency to overutilize molecular testing because it is new and physicians don’t know they don’t need it,” says Kant.

But the lab may also be called upon to explain a test. “There are a number of new tests offered through niche labs that may be complex and proprietary, and sometimes the in-house lab needs to provide information to the physician,” says Kant.

The lab can also provide education about samples that are sent out. “Physicians may order a full panel when they can start with just a few tests, which is less expensive,” says Kant.

Similarly, labs are called on to explain results, some of which are more straightforward than others. “Some tests don’t have well-established terms of significance so the lab adds value,” says Roche’s Floberg. Some labs require that the physician uses a genetic counselor in interpreting the results or providing them to patients.

“There is a tendency to not provide interpretation of a result so that it is meaningful, but if we provide as complete a picture as possible and assert ourselves as members of the team, providing the best results, people will call for information and recommendations, and our potential role becomes greater,” says Kant.

Wayne W. Grody, MD, PhD, FASCP, University of California, Los Angeles

“The lab must take on much responsibility in providing physician education. We include a lot of material with our results, not just one line,” says Wayne W Grody, MD, PhD, FASCP, professor, division of medical genetics and molecular pathology, departments of pathology and laboratory medicine, pediatrics and human genetics, at the University of California, Los Angeles (UCLA) School of Medicine.

In-Lab Development
The inclusion of detailed individual reports can add time to the turnaround, increasing the importance of having trained staff, notes Grody. Staffing the molecular diagnostics lab, particularly if needed 24/7 as in a hospital setting, is as challenging as in any other lab, sometimes more so. Because the field is still young, the number of experienced technologists is fewer than in other areas. And because many tests are still developed in-house, specific expertise is required.

“Because molecular diagnostics is new and many of the tests are esoteric without huge markets, much testing is done with laboratory-developed assays,” says Grody. Also known as home brews, these tests are more important for some labs than others.

“It depends on the technical expertise available. It is not as easy to develop these tests without some level of research knowledge, both technique and theory. Home brews save money over commercial platforms, but also, many times, the assays are not available for low-volume tests. If you want a broad menu, you must use home brews,” says Kant.

Laboratory-developed assays introduce issues of validity and standardization. “The lack of standardization in home brews and assays doesn’t mean the quality is bad, but because techniques vary across the country, it’s difficult to relate results. The industry is trying to address this issue with national guidelines,” says Grody.

“If the lab is confident in the laboratory-developed assays and every aspect of them—accuracy, specificity, sensitivity—they are almost always cheaper to run than buy. But if the lab can afford a kit from a manufacturer, it can be more certain in its results,” says Farkas.

Cost of Testing
Use of laboratory-developed assays, however, does not ensure a low budget. In addition to staff costs, intellectual-property rights and gene patents may create additional expenses. “Use of some of these technologies may require royalty payments for which there is no reimbursement code,” says Kant, who notes the PCR patent just expired earlier this year.

“Reimbursement is not clearly established. Many tests don’t have an analyte-specific code, so the lab must use method codes, but sometimes the reimbursement for these has not been determined,” says Floberg.

“There is a lack of clarity with CMS [Centers for Medicare and Medicaid Services of Baltimore]. A lot of testing is paid for by payors outside of CMS and affects why molecular diagnostics labs are not seen more frequently in community hospitals,” says Jay B Jones, PhD, director for the Chemistry and Point of Care Regional Laboratories, Geisinger Health System (Danville, Pa). He believes that some facilities have made a leap of faith in offering molecular diagnostics, and that if reimbursement does not come through, they will have to decrease their offerings in this area.

“Money is always a problem, but if testing demands the equipment, it will be covered. There is some reimbursement and economic benefit tied to the value of results, with decreased hospital stays, more appropriate use of antibiotics, less development of drug resistance, better outcomes, and greater patient safety,” says Nowak.

Proving Itself
All of these benefits are possible but not certain. Some tests do not have their clinical utility established. “Almost every week, there is some new gene or disease to test. Once we know the gene sequence, it’s easy to design the DNA test, but it takes longer to establish its clinical validity. If you detect the mutation, does it predict disease?” asks Grody.

In instances where clinical utility has been proven, but a traditional test already exists, some question the value of replacing the traditional version with a molecular one. “If you can get 95 percent of the information from a traditional test, is it worth spending more money on molecular diagnostics?” asks Kant.

A third question that should be considered is the value of automation. “For reference labs with high volumes, it’s useful to automate testing, but most hospitals probably have volumes too low to justify this expense. Many labs currently process ten or fewer samples at a time. If you have to wait to batch specimens, you cannot take advantage of the rapidity,” says Nowak.

Some tests will likely remain too esoteric to justify the expense of developing something commercially, leaving the molecular diagnostics labs to continue to produce their own assays. For this reason, a centralized structure helps, providing a single point of contact that can deliver and gather information.

However, the lab should not be segregated, and some argue for decentralization. “More and more traditional methods may be displaced by molecular tests, so these processes will move into the other clinical labs,” predicts Nowak.

However slowly, molecular diagnostics will continue to expand and play a bigger role. Instruments in development will make it easier, and more hospitals will offer these services. Greg Tsongalis, PhD, director, molecular pathology, and associate professor of pathology at Dartmouth Medical School, Dartmouth-Hitchcock Medical Center, says, “The notion of molecular diagnostics as a voodoo science is gone.”

Renee DiIulio is a contributing writer for Clinical Lab Products.