More than one-third of cancer patients with contraindications to bevacizumab receive the drug, and those without contraindications frequently experience complications after taking the agent. That is the conclusion of a study now available online ahead of print in the Journal of Clinical Oncology (doi/10.1200/JCO.2012.48.4857). The researchers also identified racial disparities associated with use of the drug.
“When an agent is approved, it is done so under ideal conditions,” said the study's first author, Dawn L. Hershman, MD, MS, Leader of the Breast Cancer Program and Associate Professor of Medicine and Epidemiology at Herbert Irving Comprehensive Cancer Center at Columbia University Medical Center/NewYork-Presbyterian Hospital. “Patients are selected to be on clinical trials with strict criteria that reduce the risks of drug toxicity.”
In the clinical setting, though, these same agents are also used in patients who would not have been selected to be in these trials, she continued. Oncologists are then left to determine the risk-benefit ratio in the community setting. Bevacizumab can result in improved survival in some diseases such as colon cancer, but in breast and lung cancers, the benefits aren't so clear, she said.
Asked for his opinion, Eric Bernicker, MD, FACP, an oncologist in the Department of Thoracic Medical Oncology at Houston Methodist Cancer Center, said the study highlights two issues: The first is whether patients receiving the drug really need it. The second is that the complication rate in those who did receive it was relatively high: “This speaks to the fact that with targeted therapies, patient selection matters,” he said.
The study used data from Surveillance, Epidemiology and End Results (SEER) Program-Medicare database to identify a total of 3,039 patients age 65 or older diagnosed with stage IV or recurrent breast cancer, non-small-cell lung cancer, or colon cancer between January 1, 2004, and December 31, 2007, and who had received bevacizumab. Patients were 65 and older and underwent follow-up until 2009.
Receiving bevacizumab was found to be associated with white race, later year of diagnosis, tumor type, and decreased comorbid conditions. Patients with colon cancer, who have been shown to benefit from bevacizumab, were five times more likely to receive the drug than breast cancer patients, while those with lung cancer were 1.2 times as likely.
The results showed that overall, 35.5 percent of patients receiving bevacizumab had a contraindication to the agent. Those with lung or colon cancer were more likely to have a contraindication than patients with breast cancer (odds ratios of 1.7 and 1.4, respectively).
And while African Americans were less likely to be treated with bevacizumab, when they did receive the agent, they were more likely to have a contraindication. Specifically, multivariate analysis indicated that receiving bevacizumab with a contraindication was 2.6 times more likely in African Americans than in whites. It was also common in patients of increased age, with more comorbidities, later year of diagnosis, and lower socioeconomic status.
In the group with no contraindication, 30.3 percent had a complication after receiving bevacizumab. Toxicity rates in these individuals were much higher in the clinical setting than in what has been reported in clinical trials, Hershman said. Furthermore, African Americans were more likely to have a complication than white patients were.
One possible reason that African Americans in the study may have been less likely to receive bevacizumab could be due to the quality of the hospitals at which they received care, she said.
Cheryl Czerlanis, MD, Assistant Professor in the Division of Hematology/Oncology at Loyola University Medical Center, also speculated via email that in addition to lack of access to quality care, the cost of targeted cancer therapies may have been another barrier, although the study did not evaluate these variables. Decreased use of bevacizumab in this population may also be related to patient and physician attitudes about perceived efficacy or the potential for toxicity, which could vary among racial groups, she added.
Other factors that may have contributed to increased bevacizumab toxicity in individuals without a contraindication could have been comorbidities that are common in African Americans, such as cardiovascular disease, stroke, and poor health behaviors, she added.
The data may also reflect different socioeconomic conditions or third-party health care coverage among specific patient populations, although that is purely speculative, noted Maurie Markman, MD, Senior Vice President of Oncology Services and National Director of Medical Oncology at Cancer Treatment Centers of America.
Overall, the study indicates that “we have to be careful when giving drugs with a lot of side effects to patients who may not fit the strict eligibility of clinical trial criteria,” Hershman said.
At the same time, oncologists want to give patients with metastatic disease any therapies that might help, and a better treatment response may be possible with bevacizumab, she continued. Physicians need to determine who may be at risk of side effects and monitor them carefully or offer prophylaxis ahead of time.
Czerlanis noted that choosing a cancer treatment, of course, is complex, intricate, and influenced by both patient and physician attitudes and behaviors. Because promising efficacy has been demonstrated in specific patient populations, physicians may infer that a benefit would extend to comparable groups. For the promise of efficacy, physicians and patients may be willing to accept the potential for increased toxicity even with a contraindication.
Still, said Markman, the term “contraindicated” is strong. While mistakenly deciding to employ bevacizumab in the management of a specific individual is possible, “an alternative view would be that what is being observed in this analysis is the tremendous complexity of caring for the cancer patient.”
The presence of one or more additional relevant medical conditions, for example, prior heart attack or known vascular disease, at the time of cancer diagnosis is common. This is particularly true in the elderly, the focus of this report, he said.
And while further evaluating the nature of the contraindications noted in this study is important, “it is highly likely that one will find oncologists attempting to do their best to help patients in the absence of optimal data because of the serious limitations associated with evidence-based Phase III randomized trials,” he said.
A number of explanations probably exist for why physicians are prescribing agents with contraindications, Bernicker said. “We need to better understand prescribing practices and ensure that quality of care continues to be an issue.”
Clinical Trial System
The Hershman et al analysis represents a “critically important example of an extremely serious problem with the current oncology clinical trials system, where patients with common and relevant comorbid medical conditions are highly likely to simply be excluded from the studies,” Markman said via email.
Consequently, when a new agent leaves the domain of limited access within trials to be employed in routine clinical practice, there are often few data that clinicians can rely upon to know how to optimally and safety administer the drug. “It is essential that this rather absurd and potentially dangerous situation is radically changed,” he said.
Oncologists are making the assumption that if a drug works in a clinical trial setting, it works universally, Hershman said. “We need to conduct clinical trials that are more generalizable.”
Bernicker agreed: While clinical trials test drugs in highly selected patients, folks in the real world don't fall into this category. They tend to have comorbidities and are older.” Clearly, clinical trials need to enroll older Americans. Currently, however, most elderly patients in the study setting tend to be fit, he said.
In addition to results often not being generalizable, prescribing physicians may not be fully aware of the specific eligibility criteria utilized in relevant clinical trials, Czerlanis added. “In the last few years, select journals have started to require the protocol for publication of clinical trial results, which is particularly useful for understanding adverse event management,” she said.
Bernicker pointed out, though, that eligibility requirements may be especially strict because the pharmaceutical companies that fund studies want to demonstrate the most positive drug benefits possible so their agents come to market—“and this does makes sense because if drugs don't come to market, we won't have them to use.
“Still, we need to continue postmarketing studies as drugs permeate out into a wider circle of patients to track efficacy data and toxicity.”
When to Use Bevacizumab
In the study, the heaviest use of bevacizumab was in lung and colon cancer patients, which is where it should be, he said.
A good rule, said Czerlanis, is to consider the patient populations treated in clinical trials evaluating bevacizumab, and “if a patient for whom a physician is considering giving bevacizumab falls within the eligibility criteria for the trial, reasonable inferences can be made about the potential efficacy and toxicities in similar populations.”
And, concluded Markman, as is the case with essentially all anti-neoplastic therapy, bevacizumab can produce serious toxicities, so patients receiving the agent need to be carefully monitored for side effects.”