The link between malignancy and thrombosis is hardly new-the two have been yoked together in the minds of physicians since Trousseau’s observations in the mid-19th century.
What is new are reinvigorated efforts to understand exactly how this link is forged, and what it means for patients. With the recent CLOT trial (Lee AYY, et al. N Engl J Med. 2003; 349: 146-153), a study on the potential for low-molecular-weight heparin (versus an oral anticoagulant) to prevent recurrent thrombotic events in cancer patients, those efforts have received another shot in the arm.
For pathologists, curiously, the CLOT trial may offer a certain respite. Hard to imagine-is it possible the coagulation maze will become easier to run?-but possible nonetheless. Because low-molecular-weight heparin rarely requires monitoring, if recent research stands up to further study, laboratories may eventually see widespread use of LMWHs and a declining need for monitoring.
Indeed, there’s nothing inherently difficult about monitoring LMWHs, says Rodger Bick, MD, PhD, clinical professor of medicine and pathology, University of Texas Southwestern Medical Center, and director of the Dallas Thrombosis-Hemostasis Clinical Center. They cannot be monitored with the activated partial thromboplastin time, as regular heparin is, but rather must be followed with anti-factor Xa levels. But, as Dr. Bick points out, as of 1998, all CLIA-certified laboratories in the United States are required to correlate their APTTs with anti-factor Xa levels. Moreover, only certain patient populations need to be measured at all: pregnant women, children, obese patients, and those with impaired renal function.
"For pathologists, the primary implication [of the latest LMWH research] is that they’re going to see more and more patients-cancer and otherwise-treated with low-molecular-weight heparins, from a wide variety of manufacturers," says Dr. Bick, who wrote a "Perspective" article that accompanied the recent CLOT study (N Engl J Med. 2003; 349: 109-111).
Does this mean pathologists can simply sit back and relax as others do the heavy lifting of sorting through malignancy and thrombosis? Certainly labs aren’t in the center of this particular storm-but they may want to keep an eye on it as it unfolds, like viewers marking a hurricane as it blows across The Weather Channel.
Those who study malignancy and thrombosis resemble genealogists doggedly pursuing family connections. There’s nothing simple about this particular family tree, however. Picture branches created by questionable matches between cousins; suspected liaisons; and too many people sharing the same name-you get the idea.
"Basically, no one has a very good handle on it," says Robert Wolff, MD, associate professor of medicine, Department of GI Medical Oncology, University of Texas MD Anderson Cancer Center.
The starting point is that cancer patients are procoagulant, at increased risk for developing thromboses due to their primary tumors and from the chemotherapeutic agents they receive. Historically, says Steven Deitcher, MD, adenocarcinomas of aerodigestive tract etiologies were thought to be most closely linked to thrombosis. "We now have a better appreciation of the fact that almost any histology of systemic cancer can promote blood clotting, whether it’s by a direct biochemical means or solely by the fact that a large, say, tumor of the pelvis, compresses blood vessels."
Additional research indicates biochemical mechanisms of coagulation "may actually be important in a tumor’s evasion of the host immune system, vascular invasion, angiogenesis, and metastasis," says Dr. Deitcher, head, Section of Hematology and Coagulation Medicine, Department of Hematology and Medical Oncology, Cleveland Clinic Foundation.
It has also become evident that patients with certain histologies, at certain stages, may benefit from receiving anticoagulant therapy. "And then the thought started being, ’Maybe the anticoagulants are interfering with the mechanisms that the tumor uses to facilitate invasion and metastasis.’ Maybe," says Dr. Deitcher, "blood thinners are anti-invasion, antimetastases-type drugs as well as being anticoagulants."
In short, the one-way street that was once thought to connect malignancy and thrombosis is retreating in the rearview mirror, and researchers are now pulling onto a multilane freeway as they explore the whys and hows of who’s at risk and who might benefit from what treatments-and who will pay for it.
Though it’s clear that patients with certain malignant diseases have elevated markers of coagulation activation, it’s not clear that these markers reflect who’s at risk for developing thrombosis, says Dr. Deitcher. "That’s because different tumors in different patients of a different stage, of a different grade, of a different distribution, probably affect coagulation differently," he says. So much for ordering one or two blood tests to predict who’s at risk of clotting. On the other hand, patients with higher levels of these activation products appear to have more progressive disease and shorter survival, he notes.
Naturally, there’s plenty of interest in developing new antithrombotic agents, with an eye toward optimizing treatments of blood clots in cancer patients when they do form. On the flip side, researchers are also staking out means of preventing clots from forming. Other researchers are trying to sort out exactly which subfraction of LMWH may have antiangiogenic effects. And drug makers have turned their attention to developing oral forms of LMWHs.
Dr. Deitcher and his colleagues are looking at the effect of giving cancer patients low-and, in some cases, high-doses of anticoagulants. "We’re starting to appreciate that you probably have to catch the patients earlier than we once thought-you can’t just sort of step in when they’re months away from dying and think that you’re going to have a significant impact on the duration of their life just by starting blood thinner," Dr. Deitcher says. "It’s been shown you probably have to start early, before they metastasize."
He and fellow researchers are also looking into whether development of clots in subjects with cancer imparts a worse prognosis-and not just because of the clot itself and the potential for deep venous thrombosis or pulmonary embolism. "We’re trying to confirm an observation we’ve made that developing a blood clot itself causes the release of factors such as VEGF." Noting that tumors can cause distant clots, Dr. Deitcher says, "We’re trying to close that loop and demonstrate in our animal model that getting a clot itself now makes the tumor grow faster."
Their ONCENOX study, presented in abstract form at this year’s American Society of Clinical Oncology annual meeting, noted a trend toward reduced recurrent venous thromboemboli in patients with primary intracranial malignancies who were treated with the LMWH enoxaparin. The ONCENOX and CLOT trials, along with the LITE trial (published as an abstract: Hull R, et al. Blood. 2000; 96: 449a), essentially have shown the efficacy of treatment with LMWH, Dr. Deitcher says, though, due to the differing sizes of the trials, only the CLOT trial is considered to be statistically significant. "But they all add up to showing that LMWH in a cancer patient population is well tolerated, with good compliance, and that these patients had lower thrombosis recurrence rates," he says. "So we’re starting to appreciate that a patient population that is a nightmare to monitor and treat with one antithrombotic can actually do better on a different antithrombotic agent, one that doesn’t even require monitoring."
Shaker Mousa, PhD, professor and director of the Pharmaceutical Research Institute at Albany (NY) College of Pharmacy, reports that he and his colleagues are, among other things, developing animal models that combine thrombosis and cancer in the same animal-"which has never been done before," he says. "Usually we study thrombosis in thrombosis models, and tumor in tumor models."
He’s also looking into the reason why heparin appears to improve chemo-responsiveness, which has been noted in a study by Lebeau, et al (Cancer. 1994; 74: 38-45). In Dr. Mousa’s opinion, "Even short-term use of low-molecular-weight heparin with chemo would enable you to enhance the uptake and permeation of the chemo." Besides being an anticoagulant, heparin is an anticancer agent, even independent of its anticoagulant effects, he contends. "There’s an antiangiogenesis effect of low-molecular-weight heparin," Dr. Mousa says. "The main mechanism for heparin inhibiting cancer is release of tissue factor pathway inhibitor, which is an indigenous inhibitor for tissue factor that basically controls tumor growth, tumor angiogenesis, and tumor metastases."
Dr. Mousa speaks about the potential implications of malignancy/ thrombosis research with the passion of a religious convert. In his view, clinicians still sometimes fail to connect the dots between malignancy and thrombosis, though he can appreciate the reasons for the lapse. "The cancer problem as such is overwhelming," he says. "And oncologists want to deal with the cancer problem; they don’t want to deal with the thrombosis problem-and if they do, they consider it the hematologist’s problem."
He’s not alone in his views. Based on a recently completed survey trial that he plans to present at next year’s ASCO meeting, Dr. Deitcher says, "In general, I can conclude that in most things related to cancer and coagulation, there is no consensus" among U.S. oncologists.
"We have physicians who believe strongly you should try to prevent all blood clots," he says. "We have physicians in the United States who don’t think it’s a major problem or issue. A third of the physicians who responded to this survey completely said they do think anticoagulants have the potential to impact survival and outcomes of cancer patients; two-thirds don’t believe it."
Dr. Wolff was part of an international group of physicians who looked at worldwide practices in terms of prophylaxis for and treatment of thromboembolic events in cancer patients. Their study appeared in The Oncologist (Kakkara AK, et al. 2003; 8: 381-388). "For patients undergoing cancer surgery, prophylaxis is given fairly routinely-at least half the time," Dr. Wolff says. When physicians decide against thromboprophylaxis, it’s usually because the patient is not thought to be at a high risk for developing a PE or DVT, or because the physician is concerned about the increased risk of bleeding.
One area where North American doctors are prone to routinely give thromboprophylaxis is to patients with a central venous catheter, he says. But it remains an area that is still often overlooked, says one observer.
"There is data to suggest that you can prevent these catheter-related clots from happening, particularly in high-risk patients," says Michael Linenberger, MD, medical director of apheresis and cellular therapy, Seattle Cancer Care Alliance, which is part of the University of Washington’s Fred Hutchinson Cancer Research Center. Likewise, he observes, growth factors-often used to support white blood cell count postchemotherapy or to stimulate peripheral blood stem cell mobilization-have been associated with increasing the risk of catheter-associated thrombosis, as are malpositioned catheters. "One milligram a day of warfarin can significantly reduce the chance of catheter-associated clotting complications," says Dr. Linenberger, adding that low-dose LMWH also appears to be efficacious.
Another dilemma, Dr. Mousa says, is that cancer patients have bleeding problems to begin with. For clinicians "to think about putting an anticoagulant in a patient who bleeds, that’s an equation for disaster." Yet heparin and low-molecular-weight heparin do not make the bleeding problems worse, he and others insist. "So actually there is an element of fear on the oncologists’ part," Dr. Mousa says.
It’s a fear that needs to be overcome, he continues. "There are enough clinical data and experimental data jumping in our face that we can’t walk away from them," Dr. Mousa says. The bottom line: "Heparin and low-molecular-weight heparin still have the potential of improving survival in cancer patients, independent of antithrombotic effect." Clots and tumors both depend on hypercoagulation as a feeding mechanism, he says. "If we prevent and we treat hypercoagulation, I can tell you from my experimental investigations, we are actually going to shut down that tumor," he adds. Heparin and LMWH are not just your father’s anticoagulants-they are, he suggests, antineoplastic and anti-inflammatory agents as well.
Even the most committed researchers temper their enthusiasm, however. They may be looking far ahead, but their heads are above the clouds, not in them. "What we really need at this point is translation of all the hypotheses and laboratory observations into clinical trials," says Dr. Deitcher.
And Dr. Mousa readily acknowledges the costs associated with LMWH. On a recent grand rounds, he says, he was able to convince his medical oncologist colleagues of the value of using LMWH in certain cancer patients. "But they said, ’Who’s going to pay for it?’ It’s a dilemma," he concedes.
LMWH is currently not reimbursed by Medicare, not to mention most insurers. "Sometimes we can convince insurance companies to pay for it, and we write letters and so forth," says Dr. Linenberger. "But to have a patient pay out-of-pocket, it’s $2,000 a month."
On the other hand, "There’s no question that our cancer patients have a higher rate of recurring thrombosis than noncancer patients on warfarin after their DVT," he continues. "And we know that our cancer patients are being readmitted to the hospital much more frequently for those recurrent events, because they’ve broken through their warfarin and maybe they’ve had bleeding complications."
With that in mind, he and his colleagues are analyzing data based on overall hospital costs, not just the cost of the drug and laboratory monitoring. "One hospital day would pretty much pay for a month of outpatient low-molecular-weight heparin," he says.
Compared to the clinical picture, the lab’s role is relatively uncomplicated.
Jawed Fareed, PhD, professor of pharmacology and pathology and director of the hemostasis and thrombosis research unit at Loyola University Medical Center, Maywood, Ill., returns to the fundamental link between cancer and thrombosis. "Cancer is not just cancer. It’s also a disease that can affect vascular and blood cells and coagulation. So when we have a patient who has cancer, we should also look for both the vascular and thrombotic disorders," says Dr. Fareed, a frequent research collaborator of Dr. Deitcher’s.
That link has been primarily forged on a symptomatic basis, he says-a hit-and-miss proposition at best. "So if you have a cancer patient, you better be careful, and make sure that you do additional coagulation tests on the patient to see if his coagulation system is off-balance, such as identification of thrombin generation and thrombin formation, and tests such as tissue factor released from tumor cells."
Though many point to the monitoring-free status of LMWH as one of its beauties, Dr. Fareed urges caution. "Cancer patients often have renal clearance problems, so they can accumulate the drug and bleed. So if the new form of therapy is used, monitoring of these drugs in cancer patients is crucial, as far as I’m concerned. Pathologists should consider that to be an important factor in their practice."
Dr. Bick points to guidelines from the North American and international consensus groups on antithrombotic therapy, which "basically say you need a platelet count daily or every other day while the patient is on any heparin-low-molecular-weight or regular heparin. So I would say frequent platelet counts need to be done to ward off heparin-induced thrombocytopenia, which is a disaster. Fortunately, it’s very rare," Dr. Bick says. "But it’s an absolute disaster when it occurs. And most clinicians forget to do it [the platelet count], so it’s kind of nice if pathologists remind them."
The challenge long-term will be for pathologists to identify which tumors have a greater propensity for generating thrombotic mediators, says Dr. Fareed. "This will take some time," he acknowledges, and will doubtless rely heavily on data from tumor registries.
Until more studies are done, says Dr. Linenberger, thromboprophylaxis will be decided case by case, based on a patient’s underlying risk. "In terms of my colleagues who take care of advanced GI or GYN cancers, they certainly don’t routinely put their patients on prophylaxis unless those patients have a personal history of prior DVT, or maybe a strong family history that might make them think they perhaps have an underlying hypercoagulable state that might come into play. And even then, it’s questionable how much of an additional risk carrying a hypercoagulable state might be."
Kandice Kottke-Marchant, MD, PhD, section head of hemostasis and thrombosis, Department of Clinical Pathology, Cleveland Clinic Foundation, offers up several other considerations for laboratories.
In malignancy, patients can often have nonovert or a compensated disseminated intravascular coagulation. "So that’s one thing to look at," she says, "or rule out as a cause for thrombosis in patients with malignancy."
At a recent CAP-sponsored consensus conference on thrombophilia (published in the November 2002 Archives of Pathology and Laboratory Medicine), the general recommendation "was made that if a patient presents with a DVT or a pulmonary embolism in the setting of malignancy, it’s not appropriate to do the thrombophilia testing, because in general, it’s thought that the malignancy is enough of a risk factor for thrombosis in those patients," Dr. Kottke-Marchant says.
That doesn’t exclude the possibility the patient might also have factor V Leiden, a prothrombin gene mutation, or other risk factor. "But you probably don’t need an underlying genetic abnormality plus the malignancy to give you the thrombosis," she says. "So it was thought that the thrombophilia testing is probably not necessary or cost-effective or really additive on how docs would actually treat patients with thrombosis in the setting of malignancy.
"In general, when I do my interpretive thrombophilia panel, if I get a completely negative panel, I’ll usually also make the comment that there is a close relationship between thrombosis and malignancy, and you really should exclude a malignant process in patients even if we find nothing on the thrombophilia panel," she continues. "Even if they’ve already started with the lab testing, I want to make them aware of the fact that they need to think about malignancy in a patient with thrombosis."
Dr. Kottke-Marchant also calls attention to another aspect of the hemostasis-malignancy link: that circulating inhibitors are not uncommon in some patients with malignancy, especially hematologic malignancies and lymphoproliferative disorders. "You can see acquired inhibitors to factor VIII and von Willebrand factor, which can give rise to very significant bleeding problems," she says. "Some patients with solid malignancies, especially adenocarcinomas, can have circulating heparin-like anticoagulants. So that on the flip side, most malignancies are more a risk for thrombosis; yet malignancy can also be a risk factor for bleeding. Be aware that bleeding in the face of malignancy can either be an overt DIC, which you might see in a later stage of a malignancy, or it could be something like an acquired inhibitor."
For DIC, "the evaluation is fairly similar to evaluating DIC in patients without malignancy." Keep in mind that nonovert DIC-what used to be called low-grade DIC-is often more associated with thrombosis, she says, "but once you get a much more aggressive DIC that’s not compensated by the liver, you can have fairly widespread bleeding problems and hemorrhagic problems. So one thing that people need to recognize is that all DIC is not the same, even associated with malignancy. But usually most patients with malignancy and fairly-at least initially-good liver functions tend to be producing enough of the clotting factors."
Finally, she notes, some malignancies are strikingly associated with DIC-usually the acute leukemias, especially acute promyelocytic leukemia. “A hallmark of APL is the development of a very fulminant DIC,” Dr. Kottke-Marchant says. “Screening for DIC should be part of standard medical practice in any patient that presents with an acute leukemia."
Fortunately, she concludes, "I don’t think there are a lot of traps that one falls into with testing in patients with malignancy.”
Karen Titus is CAP TODAY contributing editor and co-managing editor.