VTE risk in patients with cancer often overlooked understudied

Amid the myriad concerns that accompany a cancer diagnosis, the risk for venous thromboembolism may be last on the minds of patients and clinicians.

However, substantial data demonstrate the need to be aware of — and prepare for — this potentially deadly complication.

One in every five VTE cases occurs among patients with cancer. The incidence of VTE in this population is as high as 20%, with greater incidence among those who are hospitalized, have metastatic disease, and undergo chemotherapy or surgery.

VTE leads to a two- to sixfold increased risk for death among patients with cancer. It also may delay receipt of chemotherapy, worsen quality of life and increase health care expenses.

Although life-threatening, experts say VTE often is managed poorly, contributing to worse outcomes.

Some clinicians may administer chemotherapy without considering the risk for clotting, or they give aspirin rather than an anticoagulant to prevent clotting, according to Agnes Y.Y. Lee, MD, MSc, FRCPC, professor in the department of medicine at University of British Columbia and director of the thrombosis program at Vancouver General Hospital.

“This is an area that physicians have overlooked and it has a very paternalistic perspective,” Lee told HemOnc Today. “We assume too much without educating and asking patients.”

Patients often receive warfarin or direct oral anticoagulants (DOACs) for VTE, but data have shown that low–molecular-weight heparins (LMWHs) are a more effective alternative to warfarin, even if they are more costly and require a daily self-injection. However, a lack of data exist comparing LMWH and DOACs for patients with cancer, Lee said.

“We would never tell a patient with diabetes to continue taking a pill when they need insulin; yet, for whatever reason, for cancer patients we are willing to overlook that lack of evidence,” Lee said. “Patients with cancer want what is best for treating their disease — they do not shy away from toxic treatments. However, we have to inform them and educate them so they can make a decision that is right for them.”

There also is a lack of consensus on VTE prevention in patients with cancer, as well as questions about biomarker identification and the role of screening for occult cancer in those with unprovoked VTE. HemOnc Today spoke with clinicians and researchers about these controversies, as well as how clinicians can determine the best course of action for patients in inpatient and outpatient settings.

Guidelines

VTE comprises deep vein thrombosis — or blood clots that form in a deep vein, typically in the leg — and pulmonary embolism, which occurs when a DVT breaks free from a vein wall, travels to the lung and blocks blood supply.

VTE treatment in patients with cancer often is complex, can delay anticancer strategies, and can elevate risk for bleeding. Consequently, evidence-based protocols for VTE risk assessment, prevention and treatment are essential.

Guidelines direct clinicians to consider each patient’s situation, accounting for factors such as inpatient vs. outpatient status, whether they are ambulatory and their need for surgery.

Cancer site also matters. Among patients with solid tumors, VTE rates are highest for those with cancers of the pancreas, kidneys, ovaries, lungs and stomach. In hematologic malignancies, VTE risk is greatest for those with myeloproliferative neoplasms, multiple myeloma and non-Hodgkin’s lymphoma.

ASCO created guidelines on VTE prevention and treatment in 2007. The society issued one update in 2013, then reviewed an additional 53 studies to create another update in 2014.

“The evidence is always changing,” Gary H. Lyman, MD, MPH, professor in the school of medicine of the division of medical oncology at University of Washington, medical oncologist in the breast cancer program of Seattle Cancer Alliance and co-chair of the ASCO guidelines update committee, said in an interview. “Although new data became available, there are a number of remaining questions that still need to be addressed, and there are a number of novel agents that have since come along that need to be considered and studied.”

Lyman — who also co-chairs ASH’s thrombosis and cancer panel, which expects to release a new guideline in 2016 — said one of the positives of the latest research is that guidelines from ASCO, the National Comprehensive Cancer Network and the European Society for Medical Oncology are in agreement with how to approach VTE prevention and treatment in most settings.

These guidelines all suggest prophylaxis is appropriate for patients with cancer who are hospitalized and those undergoing major surgery. However, controversy exists on how to best manage the majority of outpatients who often receive chemotherapy in an ambulatory setting, Lyman said.

“We know they are higher risk, but being an oncologist and hematologist, I find that we are dealing with a lot of side effects — both from the cancer and the treatments,” he said. “On the scale of toxicities and side effects, blood clots have not always been on the top of our minds, yet when we look at outpatients with cancer who are given chemotherapy, the number-one cause of death is cancer, followed by infectious complications and blood clots.”

Ambulatory patients with multiple myeloma are at particularly high risk, Lyman said. ASCO recommends such patients receiving antiangiogenic agents with chemotherapy and/or dexamethasone receive prophylaxis with LMWH or low-dose aspirin.

“For patients with most other cancers — including solid tumors — the guidelines do not recommend routine anticoagulation as many of them are at relatively low risk,” Lyman said.

Despite these tools, VTE risk may not be communicated clearly to patients with cancer, Alok A. Khorana, MD, medical oncologist in the department of hematology and oncology and director of the gastrointestinal malignancies program at Cleveland Clinic’s Taussig Cancer Institute, told HemOnc Today.

Photo courtesy of Cleveland Clinic, ©Russell Lee 2013.

“Patients with cancer are woefully uneducated about the risk for clots,” Khorana said. “Clots occur in as many as one in five patients with cancer and … less than half of cancer patients know they are at risk.

“There are a lot of competing priorities when an oncologist meets with a patient,” he added. “You discuss the cancer, the prognosis, the treatment and its logistics, and expected side effects. If VTE risk is not addressed in the first or second visit, it likely will never get addressed.”

Risk assessment

VTE guidelines have evolved considerably as risk stratification improved. The Khorana score — a measure Khorana, Lyman and colleagues developed in 2008 that identifies patients at low, intermediate or high risk for VTE based on a score from 0 to 6 — accounts for variables such as cancer type, BMI, and hemoglobin, platelet and leukocyte counts.

Specific treatment regimens also may increase risk. Patients with myeloma are at increased risk because standards of care in this setting include thalidomide (Thalomid, Celgene) or lenalidomide (Revlimid, Celgene) which, when combined with chemotherapy or dexamethasone, have been shown to increase risk for VTE. The blood clots occur in as many as 40% of these patients.

However, whether other subgroups of patients should receive prophylaxis is less clear.

“The issue really becomes, who are the most high-risk patients?” Kenneth A. Bauer, MD, professor of medicine at Harvard Medical School, medical oncologist at Beth Israel Deaconess Medical Center and a HemOnc Today Editorial Board member, said in an interview. “It’s essential to find those we should treat with prophylaxis.”

There are various strategies to accomplish this, Bauer said. One is use of the Khorana score. Another is the identification of biomarkers, which could help determine whether all high-risk patients should receive prophylaxis, and whether some patients are more susceptible to VTE.

“Biomarkers will be very helpful to stratify patients according to their risk,” Marc Carrier, MD, MSc, FRCPC, associate professor of medicine in the department of medicine and scientist in the clinical epidemiology program at Ottawa Hospital Research Institute, told HemOnc Today. “We will be able to individualize therapy based on those biomarkers.”

One candidate gene that may contribute to VTE development in patients with cancer is vascular endothelial growth factor A (VEGFA). Ferroni and colleagues demonstrated that a disease- or drug-related imbalance in the VEGFA system could disturb vascular homeostasis.

In the study, published in 2015 in Thrombosis and Hemostasis, VTE occurred in 9% of the 297 patients enrolled, with a median time to event of 3.4 months. The VEGFA-1154G/A polymorphism appeared significantly associated with VTE risk. The A allele demonstrated a protective effect in the overall population (HR = 0.21; 95% CI, 0.07-0.58), and among metastatic patients who received bevacizumab (Avastin, Genentech; HR = 0.09; 95% CI, 0.01-0.86).

D-dimer level and tissue factor expression are other potential biomarkers.

“[The D-dimer level] is not to a level where it should not be recommended to use in clinical practice,” Khorana said. “There’s been some conflicting data on tissue factor,” Khorana said. “It looks like the preponderance of the evidence is in pancreas cancers, where its expression is associated with risk for thrombosis, but not in the general population.”

The phase 3 CATCH study compared treatment with tinzaparin (Innohep, Leo Pharma) alone to initial treatment with tinzaparin followed by warfarin in 900 patients with active cancer and acute DVT or PE. A secondary analysis of this trial — presented this year at ASCO by Khorana and colleagues — showed venous compression from a mass or adenopathy (HR 2.96; 95% CI 1.8-4.86) and diagnosis of hepatobiliary cancer (HR 2.91; 95% CI 1.2-7.02) significantly increased risk for recurrent VTE.

“Based on those findings, it looks like tissue factor might turn out to be a biomarker of some interest in pancreatic cancer to predict primary clots and all cancer patients to predict second clots,” Khorana said.

Although the primary debate relates to who should receive prophylaxis in an outpatient setting, the widespread use of prophylaxis for all patients in the inpatient setting is in part based on health care policy rather than evidence, Bauer said.

“I dare say that there is controversy regarding the benefits of anticoagulant prophylaxis for all medical inpatients,” he said. “For an ambulatory patient admitted overnight for a cardiac catheterization, a nurse will likely inject the patient with an anticoagulant and the patient may wonder why they need it.

“The Joint Commission on Accreditation of Healthcare Organizations [JCAHO] has promulgated policies that hospitals have protocols for DVT/PE prophylaxis in place and implement them,” Bauer added. “This is an element that is evaluated as part of JCAHO accreditation, which is required to qualify for payment by insurance carriers including Medicare. It could be argued that across-the-board prophylaxis without regard to risk factors may not be warranted, but in practice it is very hard to achieve high rates of prophylaxis for medical inpatients without such a policy in place.”

High rates of DVT prophylaxis can be achieved by having order sets including anticoagulant prophylaxis for all medical patients with the ability to opt out if there are contraindications (ie, excessive bleeding risk), Bauer said.

“There is also evidence that standard anticoagulant options used for medical thromboprophylaxis may be less effective in inpatients with active cancer,” he said.

Role of screening

Defining which patients with cancer are at greater risk for VTE also has implications for screening.

“Screening is a challenge,” Lyman said. “The real issue is whether you screen everybody or just the high-risk group. If you do not do baseline screening and a patient develops a blood clot, let’s say a pulmonary embolism 4 weeks later after chemotherapy, you do not know if it was there to begin with and is progressing or if it’s new.”

Data from Khorana and colleagues — published in December 2014 in Thrombosis Research — showed baseline VTE screening may be beneficial for high-risk patients with cancer. Results indicated about 10% of patients at baseline had occult DVT.

“The screening helps us to sort out when a clot developed, but I don’t think anyone is ready to say that every patient with cancer getting chemotherapy should have a baseline ultrasound or screen,” Lyman said. “A lot of these things are also being picked up based on routine staging studies, like CT scans.”

Because patients with cancer are at higher risk for blood clots, researchers have investigated whether those with unprovoked VTE should be screened for cancer. Approximately 10% of patients who have an unprovoked VTE are diagnosed with cancer within 1 year.

Robertson and colleagues — in a paper published this year in Cochrane Database of Systemic Reviews — showed extensive screening for cancer in patients who had a first episode of unprovoked DVT or PE led to diagnosis of less advanced malignancies (64% vs. 20%; P = .047) and earlier cancer diagnosis from time of VTE (mean, 1 month vs. 11.5 months).

However, Carrier and colleagues conducted a study — published this year in The New England Journal of Medicine — that showed no benefit of screening these patients for cancer. The addition of screening with comprehensive CT of the abdomen and pelvis to age-appropriate cancer screening did not significantly reduce the time to cancer diagnosis (4 months vs. 4.2 months) or cancer-related mortality (0.9% vs. 1.4%).

“We used to think VTE was the first sign of cancer when it was unprovoked,” Carrier said. “But we found that the incidence of cancer detection was much lower than we originally thought — about 4%. More extensive screening doesn’t mean you are going to find more cancer. It is not necessary. If we limit screening, not only is it reassuring to patients, it is cost-saving and it assures that these patients do not have to undergo further, unnecessary testing.”

Above all else, individualized care should guide screening decisions, Lyman said.

“We generally do not recommend screening for cancer if there is a single blood clot because of a very low yield, risk from radiation exposure and cost issues,” Lyman said. “But certainly, if a patient has recurrent, unexplained blood clots, or if there’s a family or personal history, it may be appropriate to take a more in-depth look. It’s a piece of the puzzle —if you see a blood clot in an older patient who is losing weight and has other symptoms, cancer is something that should be explored.”

The CALLISTO initiative

Previously, LMWH had been the only agent extensively studied in the prevention and treatment of cancer-specific thrombosis, Khorana said.

Although deemed safe and effective for the better part of 2 decades, treatment with LMWH is challenging because it requires daily self-injections, creating a barrier for outpatient therapy.

“You are asking patients with cancer who are already suffering a lot and have all the side effects to take on a daily or twice daily injection,” Khorana said. “That has been challenging.”

DOACs — daily or twice-daily pills — were first indicated for prevention of blood clots after knee or hip surgery, as well as prevention of stroke and atrial fibrillation.

Dabigatran (Pradaxa, Boehringer Ingelheim), rivaroxaban (Xarelto, Janssen), apixaban (Eliquis, Bristol-Myers Squibb) and edoxaban (Savaysa, Daiichi-Sankyo) recently have been approved for the treatment of blood clots. However, these approvals were for VTE in general; they were not specific for the treatment of VTE in patients with cancer.

However, drug companies have been slow to pursue DOACs in a cancer setting, Khorana said.

One meta-analysis — conducted by Vedovati and colleagues and published this year in CHEST — showed DOACs may be as effective as conventional anticoagulation for the treatment of VTE in selected patients with cancer. Among 1,132 patients, the rate of VTE appeared comparable among patients treated with DOACs vs. conventional anticoagulants (3.9% vs. 6%; OR = 0.63; 95% CI, 0.37-1.1).

Still, randomized, prospective studies have been lacking due to challenges inherent in a population of patients with cancer.

“That’s changed with the CALLISTO program,” said Khorana, who will serve as chair of the initiative’s advisory council. “This is a set of clinical trials addressing both prevention and treatment of thrombosis in cancer.”

The goal of CALLISTO — a Bayer–Janssen co-sponsored initiative — is to address these questions in a population of patients with cancer using the DOAC rivaroxaban. The initiative contains a set of nine studies — seven of which are prospective clinical trials — composed of more than 4,000 patients with various cancer types.

The researchers also will assess the impact of anticoagulation on other clinical factors, such as the concomitant use of rivaroxaban with standard therapies, and the assessment of bridging strategies for patients taking rivaroxaban or LMWH when chemotherapy-induced nausea or vomiting limits a patient’s ability to take oral therapy.

“This is a wonderful opportunity for patients with cancer to have access to this new class of agents that is much more patient friendly,” Khorana said. “We may be able to move away from the daily self-injections that we’ve been requiring these patients to do so far.”

Researchers look forward to these data and others in the setting of VTE and cancer. For instance, the CDC initiated a challenge this month designed to recognize hospitals and health care systems that employ original and effective VTE prevention strategies and interventions. The goal is to identify best practices and to increase use of VTE prophylaxis among hospitalized patients, including those with cancer.

“There’s a global push to improve evidence and to see how these patients do on different types of treatment,” Lee said. “[These initiatives] will provide us with more data that we desperately need, but it will take more time to collect. But, it is a step in the right direction.” – by Anthony SanFilippo

References:

Ay C and Pabinger I. Hamostaseologie. 2015;doi:10.5482/HAMO-14-11-0066.

Carrier M, et al. J Thromb Haemost. 2011;doi:10.1111/j.1538-7836.2011.04215.x.

Carrier M, et al. N Engl J Med. 2015;doi:10.1056/NEJMoa1506623.

Ferroni P, et al. Thromb Haemost. 2015;doi:10.1160/TH15-03-0259.

Khorana AA and McCrae KR. Thromb Res. 2014;doi:10.1016/S0049-3848(14)50006-0.

Khorana AA, et al. Abstract 9621. Presented at: ASCO Annual Meeting; May 29-June 2, 2015; Chicago.

Khorana AA, et al. Thromb Res. 2014;doi:10.1016/jthromres.2014.09.016.

Lyman GH, et al. J Clin Oncol. 2015;doi:10.1200/JCO.2014.59.7351.

Lyman GH. Cancer. 2011;doi:10.1002/cncer.25714.

Robertson L, et al. Cochrane Database Syst Rev. 2015;doi:10.1002/14651858.CD010837.pub2.

Vedovati MC, et al. Chest. 2015;doi:10.1378/chest.14-0402.

For more information:

Kenneth A. Bauer, MD, can be reached at kbauer@bidmc.harvard.edu.

Marc Carrier, MD, MSc, FRCPC, can be reached at mcarrier@toh.on.ca.

Alok A. Khorana, MD, can be reached at khorana@ccf.org.

Agnes Y.Y. Lee, MD, MSc, FRCPC, can be reached at alee14@bccancer.bc.ca.

Gary H. Lyman, MD, MPH, can be reached at glyman@fredhutch.org.

Disclosure: Lee reports consultant roles with and honoraria from Bayer, Bristol-Myers Squibb, Daiichi Sankyo, LEO Pharma and Pfizer, as well as research support from Bristol-Myers Squibb. Bauer, Carrier, Khorana and Lyman report no relevant financial disclosures.

 

 

Should all ambulatory patients with cancer who have a high risk for VTE receive prophylaxis?

Thromboprophylaxis should be considered for all ambulatory patients with cancer who are at high risk for venous thromboembolism.

Patients with multiple myeloma who are treated with thalidomide (Thalomid, Celgene) or lenalidomide (Revlimid, Celgene) in combination with glucocorticoids, anthracyclines and/or melphalan have an approximately 15% to 25% risk for VTE during treatment. Although randomized controlled trials of VTE prophylaxis in this patient group are lacking, evidence from studies of myeloma therapies suggests that thromboprophylaxis reduces the risk for VTE to less than 10% when administered during chemotherapy. In patients with other cancer types, the risk for VTE can be assessed using the Khorana score, which uses readily available clinical data — including primary tumor site, BMI, hemoglobin level, and white blood cell and platelet counts — to categorize individuals into low-, intermediate- and high-risk groups. High VTE risk corresponds to a Khorana score of 3 or higher. Additional VTE risk factors that should be taken into account include extent of disease, presence of a central venous catheter and personal history of unprovoked VTE. An individualized assessment of the risk for bleeding must also be incorporated into the decision of whether VTE prophylaxis is likely to be beneficial to a particular patient.

The efficacy of thromboprophylaxis in select groups of patients with varying tumor types has been examined in a number of randomized controlled trials. The majority of these studies showed significant reduction in the risk for VTE in patients receiving thromboprophylaxis without a concomitant increased risk for major bleeding. Notably, these findings hold true only for certain cancer types, particularly locally advanced or metastatic pancreatic and lung cancers. In general, comparisons of thromboprophylaxis to placebo or no treatment have failed to demonstrate an OS benefit of VTE prophylaxis for ambulatory patients with cancer.

Clinical practice guidelines from ASCO, the National Comprehensive Cancer Network, the American College of Chest Physicians and an international consensus working group each recognize that although VTE prophylaxis should not routinely be provided to all ambulatory patients with cancer, in certain high-risk populations, prophylaxis should at least be considered for primary prevention of VTE. Decision-making discussions with patients should include careful assessment of the individual’s risks for thrombosis and bleeding and incorporate the patient’s values and preferences with regard to thrombosis prevention strategies. In addition, all patients with cancer should be informed of the risk for cancer-associated thrombosis and educated on the signs and symptoms of VTE that should prompt them to seek clinical evaluation.

 

Tyler W. Buckner, MD, MSc, is an assistant professor in the divisions of hematology and pediatric hematology/oncology at University of Colorado School of Medicine. He can be reached at tyler.buckner@ucdenver.edu. Disclosure: Buckner reports no relevant financial disclosures.

Ambulatory patients with solid cancers who are high risk for venous thromboembolism should not be treated with thromboprophylaxis.

Outpatient thromboprophylaxis (OTP) for solid cancer patients receiving chemotherapy remains controversial. Although I support OTP for patients with myeloma receiving thalidomide/lenalidomide-based chemotherapy — based on the fact aspirin or low-dose warfarin prophylaxis is inexpensive, and aspirin is equivalent to warfarin and prophylaxis-dose enoxaparin in two large randomized controlled trials — the evidence for solid cancer patients is weak.

First, we have no perfect tools to define which patients are in VTE high-risk group. The evidence that Khorana score is imperfect is emerging. For example, the Khorana score components were unable to discriminate between patients at high and low risk for VTE in the FRAGMATIC lung cancer trial. The VTE rates among 910 controls varied according to Khorana score 1 (12.9%), 2 (10.9%), 3 (11.6%) and 4 (14.3%). The researchers concluded there is a need to further develop cancer-specific thrombosis risk assessment tools to help guide targeted OTP.

In a trial by Macbeth and colleagues — composed of 1,100 patients in each arm — the VTE risks were 7.1%  among controls and 4.1% among patients assigned dalteparin (Fragmin, Pfizer), and the results do not support OTP in patients with lung cancer.

Second, OTP trials in advanced pancreatic cancer have problems.

The FRAGEM is a small phase 2b trial enrolling 62 controls and 59 patients assigned dalteparin. Researchers administered dalteparin 200 IU/kg daily for the first month followed by 150 IU/kg daily for 2 more months. The CONKO-004 trial enrolled 152 controls and 160 patients assigned enoxaparin 1 mg/kg daily for 3 months followed by 40 mg daily until disease progression.

Although the FRAGEM trial excluded screen-detected incidental VTE at study entry, the CONKO-004 trial did not. Recent studies suggest patients with advanced pancreatic cancer may have substantial screen-detected incidental VTE at diagnosis. This means some CONKO-004 trial patients could have had incidental VTE at study entry and it is difficult to interpret the published results with confidence.

Despite the recent meta-analysis of advanced pancreatic cancer OTP trials showing that 11 patients require OTP to prevent one symptomatic VTE, many uncertainties remain regarding OTP duration, type and dose of low–molecular-weight heparin (LMWH), costs and potential benefits in patients with advanced pancreatic cancer.

Third, in other patients with solid cancers and a Khorana score of 3 or greater, the OTP recommendation/guidance was based on post-hoc subgroup analysis. As we are aware that there are problems with subgroup analyses, such as multiple comparison problems, over-interpretation of noted differences, apparent interaction effects, provision of indirect evidence and being subject to bias. OTP recommendation/guidance should not be given from post-hoc subgroup analysis.

Fourth, although oncologists believe that VTE is the second-leading cause of death in patients with cancer, many OPT trials in solid cancers have shown no survival benefit despite decreasing VTE incidence in the LWMH-treated groups. Unlike other supportive care modalities — such as biweekly antiemetic prophylaxis and in-hospital thromboprophylaxis — OTP requires daily LMWH injections for several months and obviously causes patient discomfort, bleeding complications and financial toxicity.

Further research is urgently needed to develop very good, disease-specific risk scoring systems and to perform well-designed large randomized controlled trials before jumping into the conclusion that TP should be given to patients with solid cancers and a high risk for VTE.

 

References:

Gelber RD and Goldhirsch A. Stat Med. 1987;6:371-388.

Khorana AA, et al. J Thromb Haemost. 2014;doi:10.1111/jth.12725.

Khorana AA, et al. Thromb Res 2014;doi:10.1016/j.thromres.2014.09.016.

Larocca A, et al. Blood. 2012;doi:10.1182/blood-2011-03-344333.

Maraveyas A, et al. Eur J Cancer. 2012;doi:10.1016/j.ejca.2011.10.017.

Macbeth F, et al. J Thorac Oncol. 2013;8(Suppl 2): Abstract O27.02.

Noble S, et al. J Thromb Haemost. 2015;13(Suppl 2):Abstract OR130.

Palumbo A, et al. J Clin Oncol. 2011;doi:10.1200/JCO.2010.31.6844.

Pelzer U, et al. J Clin Oncol. 2015;doi:10.1200/JCO.2014.55.1481.

Tun NM, et al. Blood. 2014;124:4266.

Verso M, et al. Intern Emerg Med. 2012;doi:10.1007/s11739-012-0784-y.

Thein H. Oo, MD, FRCPE, FACP, is an associate professor of medicine at The University of Texas MD Anderson Cancer Center. He can be reached at thoo@mdanderson.org. Disclosure: Oo reports no relevant financial disclosures.