Cisplatin Linked to Higher Risk for VTEs

Treatment with cisplatin is associated with a significant increase in the risk for venous thromboembolic events (VTEs), according to a meta-analysis published online November 13 in the Journal of Clinical Oncology.

The incidence of VTEs in patients with advanced solid tumors was 1.92% in those who received cisplatin-based therapy and 0.79% in those who did not.

The risk for VTEs was significantly higher in patients who received cisplatin-based chemotherapy than in those who did not (relative risk [RR], 1.67; P = .01). In an exploratory subgroup analysis, the highest relative risk for VTEs was in patients who received a weekly equivalent cisplatin dose above 30 mg/m² (RR, 2.71; P = .02) and in patients enrolled in clinical trials with results published from 2000 to 2010 (RR, 1.72; P = .01).

In general, the risk for VTEs is 4- to 8-fold higher in cancer patients than in the general population, note the authors, led by Matthew D. Galsky, MD, director of genitourinary medical oncology at the Mount Sinai School of Medicine in New York City. A number of clinical characteristics have been associated with an increased risk for VTEs in cancer patients, including the presence of multiple comorbidities, advanced disease, and poor performance status. Cancer patients who develop VTEs also have a worse prognosis and a higher mortality risk.

Prophylaxis Risk Not Worth It

Cisplatin has been previously linked to a higher risk for VTEs, but this association has been largely based on case reports, single-group prospective studies, and retrospective analyses, the authors point out. However, a phase 3 trial that prospectively evaluated VTE rates in patients with advanced gastroesophageal cancer found that there were fewer thromboembolic events in patients who received oxaliplatin than in those who received cisplatin (7.6% vs 15.1%; P < .001) (J Clin Oncol. 2009;27:3786-3793).

"It is a well-known fact that chemotherapy can induce endothelial dysfunction and contribute to cardiovascular side effects, with both venous and arterial thromboembolism associated with cisplatin," said Grace Dy, MD, assistant professor in the Department of Medicine at Roswell Park Cancer Institute in Buffalo, New York.

Dr. Dy, who was not involved in the study, noted that the 1.92% incidence reported in this analysis is much lower than that reported in other retrospective studies/pooled analyses looking at the incidence of VTE in cancer patients. "This may be due to differences in the population analyzed (restricted to patients with tumor types where cisplatin is used and no control for other risk factors), although separate analyses of gastric/lung cancer studies by other investigators indicate much higher incidences (8% to15%)," she told Medscape Medical News.

Dr. Dy explained that if the VTE incidence is truly that low, the risk associated with prophylaxis is not worth it. "The threshold for when prophylaxis becomes relevant is very arbitrary, but based on known hemorrhage rates with prophylaxis of 10% to 20% (depending on the population), a 15% to 20% risk for VTE...at least is required to make it worth the risk of prophylaxis," she said.

This meta-analysis "probably represents an underestimate," she added. "Risk is likely additive/synergistic when agents are combined."

VTEs Associated With Other Drugs

Other drugs used in the oncology setting are also associated with an increased risk for VTEs. One recent meta-analysis found anti-epidermal growth-factor receptor (EGFR) agents to be associated with a significant increase in the risk for VTEs (Ann Oncol. 2012;23:1672-1679).

An earlier meta-analysis found that the angiogenesis inhibitor bevacizumab (Avastin, Genentech/Roche) significantly increases the risk for VTEs (JAMA. 2008;300:277-2285), as reported by Medscape Medical News. Because bevacizumab is being used increasingly in the routine treatment of cancer patients, the authors of that analysis suggest that a black-box warning be considered. However, studies of the association between VTEs and bevacizumab have had mixed results.

Study Details

Dr. Galsky and colleagues conducted a systematic review and meta-analysis to evaluate the incidence and risk for VTEs associated with cisplatin-based chemotherapy, which has not been well studied.

The 8216 patients studied were drawn from 38 randomized controlled trials and had a range of advanced solid tumors.

There was significant heterogeneity in the incidence of VTEs in patient populations receiving cisplatin, ranging from 0% to 17%. The highest incidence (17%) of VTEs was observed in a trial of 80 patients with urothelial cancers who were treated with methotrexate, vinblastine, and doxorubicin, in addition to cisplatin. In contrast, 13 trials reported no VTEs in either the treatment or control groups.

In an exploratory analysis that stratified patients by tumor type, the incidence of VTEs was highest in patients with gastric/esophageal cancer (4.84%), pancreatic cancer (2.10%), and small-cell lung cancer (1.36%).

The authors hypothesized that the incidence of VTEs reported in cancer clinical trials had increased during the previous 10 years because of an increase in the use of high-resolution computed tomography (and the resultant incidental detection of pulmonary emboli). They found that the incidence of VTEs in the 10 trials published from 1990 to 1999 was 0.69%; in the 28 trials published from 2000 to 2010, the incidence was 2.67%.

In those 28 trials, cisplatin-based chemotherapy was associated with a relative risk for VTEs of 1.72. In the 10 trials published from 1990 to 1999, the relative risk was 1.30; the difference did not reach statistical significance.

"Given the morbidity and mortality associated with VTEs in patients with cancer, our study adds further support to calls for prospective trials of cisplatin-based chemotherapy administered with prophylactic anticoagulation," the authors conclude.

They add that "improved reporting of VTEs in clinical trials should be considered to better define the excess risk of VTEs associated with new and existing therapies."

Several of the coauthors report relationships with several pharmaceutical companies, as noted in the paper.

J Clin Oncol. Published online November 13, 2012. Abstract