College of American Pathologists
Printable Version

  For molecular menus, dynamic choices


CAP Today




Dr. Jean Lopategui of Cedars-Sinai Medical Center, where the laboratory’s molecular service has grown, he says, from  “humble beginnings” and a handful of test offerings to 60 to 70 tests today.
Dr. Jean Lopategui of Cedars-Sinai Medical Center, where the laboratory’s molecular service has grown, he says, from “humble beginnings” and a handful of test offerings to 60 to 70 tests today.

October 2012
Feature Story

Karen Titus

These days, molecular pathology is less testing paradigm than litmus test. How you view it says much about where you think your laboratory, and the profession itself, needs to head next.

Are molecular methods esoteric tests, skirting the edge of research? Are they the bailiwick of reference laboratories and academic medical centers, or do they also belong in small- and intermediate-sized practices? Are they best left to the next generation of pathologists? Will they pin pathologists to overseeing specimen collection and quality assurance, while consigning complex interpretations to others? Or will they expand the role of pathologists, placing them at the helm of patient care as consultants and data directors?

The molecular revolution started decades ago, with a handful of tests. Granted, it’s hard to grasp the full meaning of any revolution in its earliest days. But now the implications are becoming clear. “Molecular medicine” is on the verge of becoming a redundant phrase—medicine is molecular.

“I think back to the days when the first PCR tests came out, and then look at where we are today, versus where we were then, and it’s just unbelievable,” says Mark Lifshitz, MD, sounding like someone who awoke one morning to find himself on Mars: How did I get here? And yet Dr. Lifshitz has been no slouch in the molecular arena—he’s director of clinical laboratories at New York University Medical Center, where, among other things, he oversees an extensive molecular testing program for infectious disease. “It’s taken a long time, but we’re finally at the point where everything’s getting ready to explode.”

The explosion will push labs one way or the other—either to do more molecular testing, or less. Sitting on the fence is not an option.

Explosions are scary—there’s a reason why people run from them (and the ensuing fireballs) in action films.

It’s no different in medicine. Seasoned observers see some physicians—pathologists as well as clinicians—as being cowed by molecular testing. In the words of one pathologist, “They make it more than what it is.”

Their fears can’t be explained away solely as a generation gap. Jean Lopategui, MD, sees plenty of hesitation among residents at Cedars-Sinai Medical Center, Los Angeles, where he’s medical director, molecular medicine and clinical cytogenetics. When he asks them what they want to learn in their molecular rotation, “A lot of them tell me they’re apprehensive,” though the number eager to learn is growing. “They’re not sure what they need to know or how much they need to study for this. They view it as a prerequisite course in biochemistry.

“It’s simpler than that,” Dr. Lopategui con-tinues.


“We don’t need to know all the details,” he says. The staples, as outlined by Dr. Lopategui, will suffice. What tests should be ordered? How should they be prioritized? What kind of tissue is needed, and how much? What is the basic genetic pathway, and what does it mean for diagnosis, prognosis, predictive response to novel targeted therapies, and molecular monitoring of disease? Dr. Lopategui calls these common questions. Dodging molecular methods until one has mastered their intricacies would be like refusing to drive until one knows how to fix the transmission.

Nevertheless, he hears a common refrain from residents: molecular may be interesting, but it’s not mainstream. He points a finger at the educational curriculum to help explain this attitude. There’s not much molecular training in residency, he says, and molecular fellowships are few. Pathologists may wonder whether molecular will land them a job and instead steer themselves toward more traditional pathways in, say, dermatopathology, renal pathology, or hematopathology.

“Yes, those will likely increase their chances of getting a job,” says Dr. Lopategui. “But molecular is going to give them an even better chance. Because it applies to every field of medicine, including general pathology.”

His message isn’t just for newly minted pathologists. It applies to those who are trying to figure out whether to expand molecular testing at their own institutions, and, if so, how.

Cedars-Sinai offers plenty of lessons in how to do that, for better and for worse. Not all of its efforts have been immediately successful, at least not in traditional terms (read: breaking even, persuading all clinicians), but Dr. Lopategui, along with his colleague Mahul Amin, MD, chair, pathology and laboratory medicine, are convinced that molecular is the right way to go.

Those who agree are finding it easier than ever to travel the molecular route. “These days everything is geared toward molecular testing,” says Dr. Lifshitz, from infectious disease to anatomic pathology to genetics.

Technology is forcing the issue. In ID, says Dr. Lifshitz, some tests are practically push-button simple, making them reasonable additions to any sized hospital. “Vendors are trying to simplify their devices. So you don’t have to have a sophisticated lab to do these tests,” he says.

Case in point, says Jay M. Lieberman, MD, is the FDA clearance of Focus Diagnostics’ Simplexa influenza A/B and RSV Direct molecular assay as a moderate-complexity test. Until recently, he says, molecular testing has been mostly limited to labs that do high-complexity testing. “The challenge was, how do you get these tests closer to the patient? How do you reach community hospitals? Can you do molecular testing in urgent care centers or emergency departments?” says Dr. Lieberman, medical director of infectious disease at Quest Diagnostics and Focus Diagnostics (which is owned by Quest).

The need for quick TATs trumped other needs, which meant PCR methods couldn’t always be used clinically despite their superior performance. But the 2009 influenza pandemic exposed the deficiencies of rapid influenza diagnostic tests that were in wide use at the time, Dr. Lieberman says.

The technical hurdle in simplifying molecular methods was the extraction step, which has been eliminated in the Focus influenza and RSV test and in its test for C. difficile. Focus makes available a number of ASRs also. “So laboratories can have their own lab-developed test to broaden their molecular menu,” says Dr. Lieberman.

Such technological breakthroughs mean the dynamic pattern of molecular testing is expanding as well as speeding up. A test may start as a sendout; as demand increases, a lab may bring it back in-house. But now labs of any size can join the fray. And even as this happens, additional tests keep popping up, like gaffes in a presidential campaign. Dr. Lifshitz compares it to a leaking boat: “You’re constantly bailing out; meanwhile, more water is coming in.” This includes areas of infectious disease where formerly no molecular testing was available, says Dr. Lieberman. “I think the future of all respiratory virus testing is molecular,” he says.

That makes sense to Dr. Lifshitz, who is starting to see ID molecular testing for certain clinical situations—like respiratory infections—offered 24 hours a day. “Just like any other test, it will be part of the workup,” he says.

This brisk pattern appears to be influencing the way medical institutions think about costs. As molecular tests move nearer to patients, clinicians can act on results more quickly. And as that happens, clinical utility becomes part of the cost equation. Cost no longer strictly means what it once meant, which may be jarring to pathologists who’ve spent the last decade or so nailing down the cost of every test and reagent, trying to squeeze them into noose-like budgets, and scrabbling for deals.

“The focus is on the big picture in a way that has never happened before,” says Dr. Lifshitz, though he’s hard-pressed to say whether this phenomenon is typical only of the New York City metropolitan area. No one is looking at lab costs in a vacuum anymore. “In the old days it would be, ‘Hey, you budgeted half a million dollars for reference lab costs, and now you’re at $600,000—why are you over?’”

“It’s not that people aren’t concerned about cost,” Dr. Lifshitz says. “But the major push has been getting information, getting patients in and out of hospitals, and reducing infection rates, and enhancing patient safety. And molecular is a part of that.” There’s less resistance to bringing new technologies onboard and ordering higher-cost tests. “People are looking at total costs associated with managing a patient, as opposed to specific costs associated with the lab.”

No one has worked out the metrics of isolating the specific contribution a test makes to the total patient experience, he says. That leaves Dr. Lifshitz to consider other factors when deciding whether to expand or contract his molecular offerings.

This being New York City, space is near the top of the list. It’s impossible to add a new test and technology platform, even if the lab is fully capable of running it, if there is no room to set it up.

There are other practical matters as well. New platforms may not run the tests the lab wants to run, or in the right combinations. There may not be enough staff to run new tests. “And sometimes it’s just a matter of time,” Dr. Lifshitz says. “Even the simplest of analyzers take time to implement. And there are a lot of competing priorities going on in hospitals these days.”

But that’s OK. Molecular can be like opera—why struggle to mount the Ring Cycle if you’ve only got the cast for “Bluebeard’s Castle”? There’s no need for labs to run every molecular test themselves, were that even possible. “You don’t need to have dozens of tests to have an impact,” Dr. Lifshitz says.

In his lab, molecular tests account for approximately 18 to 20 percent of total sendout costs. By test volume, molecular tests make up about six to eight percent of sendout tests. Parsing it further, he estimates that over a typical year the lab sends out close to 1,100 different types of tests. The 20 most costly, based on total dollars, account for about half the total sendout budget. And of those top 20 tests, some eight of them are molecular. His point in presenting these numbers is simple: Molecular tests are costly. “You don’t have to send out that many to have a significant impact on your budget.”

He continues to look for ways to bring these tests in-house—but, interestingly, not strictly as a way to cut costs. “You don’t save that much money bringing in molecular tests,” he says. It’s not like certain immunoassays, where a sendout costing $50/test can be brought in-house for $5. “The reason you do it in-house is because you get the results out quicker. Everything is tied to treatment and diagnosis, and everyone wants information quickly these days.” In these cases, clinical necessity bumps the budget aside.

That demand for quicker TATs—sometimes as part of a treatment protocol—is often what spurs the lab to bring a molecular test in-house, he says. The other nudge comes from looking at the lab’s monthly accounts. When he sees consistent, growing demand for a test, he may find it cheaper, or faster, or both to bring it in-house.

But mostly, he finds that bringing molecular tests back in has saved his lab little, and sometimes no, money. The tests are expensive, and sometimes they need to be rerun. They can also be complex, which means fewer lab personnel are trained on the system.

None of these factors will deter him from expanding his molecular menu if he feels the clinical utility is there, and if overall patient care costs drop. The profit lies elsewhere. “It’s almost like the cost of doing business,” he says. “Sometimes you have to provide information, and it costs what it costs. No one is interested in results that come back in a week.”

Dr. Lieberman echoes that view. Moving to molecular isn’t without its challenges, and he acknowledges the real concerns labs have about up-front costs. But, he says, “There’s a real cost involved when you make a wrong diagnosis. We would hope people would look at the big picture.”

“You’ve got to take the leap,” he insists.

Cedars-Sinai leapt early and continues to do so often. The laboratory offers what Dr. Amin calls a full-service molecular laboratory, providing comprehensive testing for anatomic (surgical, cytopathology) and clinical pathology (infectious disease, hematologic malignancies, familial diseases). Beyond that, the lab offers pharmacogenomics tests and array CGH for postnatal testing (with prenatal testing scheduled to start this fall). It’s also implementing next-generation sequencing, starting with a commercially available universal cancer panel of 46 genes to look at 739 selected hot-spot mutations.

Dr. Amin sees no-thing extraordinary in this. “In providing contemporary laboratory diagnostics, you have to have a molecular laboratory that is providing a relatively broad menu,” he says.

Cedars-Sinai began expanding its molecular service in 2006, when Dr. Amin arrived at the institution. The department had a clear molecular presence, he recalled, in hematologic malignancies and, to a lesser extent, infectious diseases. “It was a very strong program, but not operating under a vision and mandate to be the latest and greatest,” he says.

Dr. Lopategui is even more frank in his assessment. “We had humble beginnings,” he says.

That changed with two jolts of energy. One came from the decision to place the laboratory in the new wave, so to speak, of the profession. The second came in 2008, when the institution decided to offer personalized medicine in multidisciplinary care.

In both cases, says Dr. Amin, it quickly became obvious that the lab’s current molecular offerings would not be enough to achieve the larger vision. Every lab—no matter what the size or setting—is doing some molecular testing, Dr. Amin notes. And while Cedars-Sinai was clearly beyond the basics, it took a conscious effort to push itself even further.

Cedars-Sinai also brought in a molecular genetic pathology fellowship. If the laboratory wanted to speed ahead with the next generation of pathology leaders, Dr. Amin reasoned, the starting line would need to be filled with residents and fellows who understood molecular pathology.

Newly invigorated, the laboratory’s molecular offerings have grown from a handful of tests to something more along the lines of a California Pizza Kitchen menu—60 to 70 tests, Dr. Lopategui estimates.

Like producing lines of poetry, success hasn’t come without pain. A closer look at the incipient pharmacogenomics program—yes, it’s still incipient despite several years in existence—provides an unvarnished view of what can happen between envisioning bold new steps and actually stepping into the clinical hallways.

Dr. Lopategui, who heads up the pharmacogenomics testing program at Cedars-Sinai, is an unabashed supporter of such testing. He harbors no doubt that pharmacogenomics testing is valuable. And yet he and his laboratory colleagues can’t always get the clinicians to agree, especially when it comes to using it to guide dosing of warfarin and Plavix.

Their resistance takes the form of several arguments. One is the lack of data. The evidence of its benefit is controversial, Dr. Lopategui concedes. Without ironclad data, practice guidelines will be in short supply. Without such a consensus, clinicians are less convinced.

That adds to the burden on pathologists to educate clinical colleagues about a test’s value. “The bottom line in tests that we think are a medical necessity, but where the incontrovertible data are still not available, is to have guidelines, best practices, documents that the College can provide as resources to pathologists, who can show them to clinicians,” says Dr. Amin. Pathologists need to be frank about the materials, he says, saying that the data points in a certain direction while acknowledging more research is needed. “But then we can have these discussions with clinicians.”

In a sense, it’s a way of marketing to clinicians, without the pricey dinners that conventional sales folks might use to, ahem, convince clinicians that a test is worthy. These aren’t necessarily new issues. Any new test, especially if it’s coming from the lab and not a steak-infused clinician, requires some work to overcome what Dr. Lopategui calls the force of habit.

In the case of warfarin testing, the habit of using INRs to guide dosing stretches back to the Eisenhower era. Despite strong evidence showing that pharmacogenomics testing would help clinicians find the right dose faster than repeated INRs, clinicians have been reluctant to make the switch.

Acceptance may be growing. In 2007 the FDA issued a first label warning about warfarin sensitivity in CYP2C9 poor metabolizers and in 2010 updated its label to include the effect of VKORC1 and provide genotype-guided dosing ranges. But it might still seem like a risky venture for labs to be offering this type of molecular testing.

In that case, it was even more risky some six years ago, when Cedars-Sinai launched its venture. What spurred them to take the plunge?

Dr. Lopategui cites the enormous costs of adverse drug events. If pharmacogenomics testing could identify how patients would respond to a dosage or treatment, he says, using it seemed like a no-brainer.

But clinicians weren’t convinced. “I didn’t expect so much resistance,” he says.

“It was probably too early to launch these tests,” he continues. “We didn’t think it would take so long for the evidence to be established. We felt, from our own experience in-house and from our patient population, that this test was useful.” The lab started with warfarin testing, concluding it would be of value after a study involving 100 of its own patients. “But across the nation, that didn’t turn out to be a consensus.”

Nevertheless, Cedars-Sinai refuses to retrench.

“We’re going to continue to offer this test,” Dr. Lopategui says. “We’re still standing our ground, despite low volumes. In our experience over the last four years, there is no question this is a helpful test.”

And to be clear, this is one sliver of the molecular testing program. Molecular tests for ER, PR, HER2, EGFR, KRAS, and BRAF—to name just a few—have been much easier sells. And last year the FDA approved the first pharmacogenomics companion diagnostic test to select vemurafenib for treatment of metastatic or unresectable malignant melanoma. That’s why, despite any weariness stemming from the warfarin experience, Dr. Amin has not lost heart. Molecular medicine is already mainstream, he says—no one is waiting for an announcement to make it so. Pathologists need to do nothing less than embrace it, he says, or they’ll face dire consequences.

Next month, a look at how smaller practices are taking on molecular.

More than anything, he says, pathologists need to make sure to present molecular tests as professional interpretations. Pathologists are integrating clinical, morphologic, and phenotypic data into molecular results. Woe to those who lose sight of that and think of molecular as a mere commodity. “We will have lost the cause,” says Dr. Amin. “I can assure you that every five years we are going to lose a large chunk of what we do professionally.”

“Thinking ahead three, five, 10 years requires a vision,” he says. “But sometimes in trying to muddle through the day’s problems, we miss big opportunities.”

Among those opportunities, he says, is an expanded professional role. In the molecular world, pathologists can help clinicians sort through test options and develop best practices. Clinicians can be pack rats—bringing on a new test doesn’t mean they’ll jettison others. “Dropping tests has not been part of the process,” says Dr. Amin, diplomatically.

It will only get worse as more information becomes available. Looking at next-generation sequencing as an example, Dr. Lopategui notes that labs will be generating more information that’s also more accurate. “What do you do with 3.2 billion points of information?” he asks.

That’s where the pathologist consultant can step in, Dr. Lopategui says. As bioinformatics experts, they’ll be able to make judgments based on databases, history, and tumor classifications to decide whether the mutation is pathogenic, whether it can be treated with an available agent, or whether it will make a patient eligible for a clinical trial for a new compound.

But without molecular information, and the will to act on it, pathologists will have little to offer, fears Dr. Lopategui.

Would it were done quickly. As even Macbeth learned, dithering doesn’t work. Pathologists who proceed no further in this business will soon find themselves with an almost insurmountable knowledge gap. Says Dr. Lopategui: “If they don’t do it now, they’ll never catch up.”

Karen Titus is CAP TODAY contributing editor and co-managing editor.

Related Links Related Links