BEVACIZUMAB SAFE IN NON SQUAMOUS NSCLC

July 20, 2010 – Toxicities associated with bevacizumab (Avastin) appear to be manageable when the drug is delivered as first-line therapy with various chemotherapy combinations to patients with nonsquamous nonsmall-cell lung cancer (NSCLC), report investigators from the international postmarketing study dubbed SAiL (Safety of Avastin in Lung).

In an uncontrolled open-label phase 4 study designed to reassure clinicians about the safety of bevacizumab, adverse events with the angiogenesis inhibitor were comparable to those seen in phase 3 trials, and no new safety signals were seen, write Lucio Crinò, MD, from the Department of Oncology, Hospital Santa Maria della Misericordia, in Perugia, Italy, and colleagues. The study was published online July 20 in the Lancet.

"The study findings confirm the well-established and manageable safety profile of bevacizumab, add to the efficacy evidence supporting the use of bevacizumab in first-line treatment, and establish the relevance of findings from randomized phase 3 studies to treatment with bevacizumab in a clinical-practice setting," they write.

The data are reassuring, agrees Robert Pirker, MD, from the Department of Medicine I at the Medical University of Vienna, Austria, in an accompanying editorial.

"First, careful patient selection according to the criteria used in SAiL and surveillance during treatment should guarantee the safe use of bevacizumab in routine practice," he writes. "Second, bevacizumab seems to be safe in patients with brain metastasis. This finding is relevant because brain metastases are present at diagnosis or will develop during the course of the disease in many patients. Additionally, brain imaging before treatment with bevacizumab is no longer required."

And in the "no news is good news" category, the absence of any unforeseen toxicities with bevacizumab in the trial suggest that the drug is safe in a broad patient population, Dr. Pirker adds.

However, although this study demonstrates the safety of bevacizumab in the clinical setting, it does not add to the body of evidence of the targeted agent's efficacy in NSCLC, he notes.

"Although safety concerns about bevacizumab have now been sufficiently resolved, the debate about the drug's effect on survival continues," he writes.

"SAiL, as a single-group study, does not provide a definite answer because outcomes in patients with advanced NSCLC depend on their baseline characteristics. Thus, only randomized trials allow the effect on survival to be established," he explains.

SAiL Setting

Investigators in 40 centers in Europe, Canada, China, and Australia enrolled patients with untreated stage IIIB to IV nonsquamous NSCLC. To be eligible, the patients were required to have histologically or cytologically documented inoperable disease, an Eastern Cooperative Oncology Group performance status from 0 to 2, and adequate liver and kidney function and hematologic parameters.

The patients were treated with standard chemotherapy regimens chosen at the discretion of the investigator, plus bevacizumab at a dose of either 7.5 mg/kg or 15 mg/kg (also at the investigator's discretion) every 3 weeks for up to 6 cycles, followed by maintenance with single-agent bevacizumab until disease progression. A large majority of patients (88%) received the higher bevacizumab dose.

A total of 2212 patients, enrolled from August 2006 to June 2009, were included in the intention-to-treat analysis. The median age was 58.8 years (range, 24 to 86 years).

The primary end point was safety. Secondary end points were efficacy measured as time to disease progression and overall survival, and safety among patients who developed central nervous system (CNS) metastases during treatment or up to 6 months afterward (confirmed CNS metastases were an exclusion criterion, but CNS imaging was not required for enrollment, so some patients with undiagnosed metastases could have been enrolled, the authors noted).

About half of all patients (49%) received bevacizumab with a carboplatin doublet, 38% received it with a cisplatin doublet, and 13% received nonplatinum doublets, monotherapy, or switched chemotherapy triplets or quadruplets.

Safety Results

The overall incidence of any adverse event in the intention-to-treat population was 38%, and 13% of all patients had a grade 3 or 4 adverse event attributable to bevacizumab. Thromboembolic events of any grade occurred in 6 of patients, bleeding in 4%, and gastrointestinal perforations in 1%. Respiratory, mediastinal, and thoracic disorders of grade 3 or greater occurred in 3% (including pulmonary emboli in 1%); gastrointestinal disorders, blood and lymphatic disorders, and vascular disorders in 2% each; and nervous system disorders in 1%.

Clinically significant hemorrhaging occurred in 15 patients, fatal pulmonary hemorrhage occurred in 2 patients, and fatal hemoptysis occurred in 6.

Grade 3 or greater neutropenia was seen in 6% of patients and febrile neutropenia was seen in 3%. Grade 3 or greater nausea and vomiting each occurred in 3% of patients.

A total of 281 patients were determined to have CNS metastases on study, and of these, 5 had CNS bleeding; the decision to discontinue treatment was left to the treating physician. Another 7 of these patients had CNS metastases detected after disease progression and were therefore excluded from further analysis.

"The eligibility criteria for SAiL allowed enrolment of a broad range of patients, including those receiving concomitant treatment, elderly patients, and patients with a performance status of 2. Such patients are usually underrepresented in clinical trials, mainly because of an increased prevalence of comorbidities. The findings of this study provide reassurance that, with appropriate management, all of these patients can benefit from bevacizumab-based therapy without an increased risk of bevacizumab-related serious adverse events," the investigators write.

Efficacy Analysis

The secondary efficacy analysis showed that, as of the final data cutoff of July 24, 2009 (mean follow-up 380.4 days), 1525 patients (69%) had progressive disease and 723 (33%) were alive. The median time to progression was 7.8 months (95% confidence interval, 13.8 - 15.3 months). Patients who received monotherapy or a nonplatinum agent had slightly lower median overall survival; otherwise, efficacy was similar across chemotherapy regimens. Disease control rates among 2036 patients with disease assessments after baseline were 3% for complete response, 48% for partial response, and 37% for stable disease.

The study was funded by F. Hoffmann-La Roche Ltd. (Roche), which markets bevacizumab. Dr. Crinò and 9 of his coauthors report receiving honoraria, travel grants, research support, or other remuneration from the company. Dr. Pirker reports receiving speaker's fees and honoraria for consulting from AstraZeneca, Eli Lilly, Merck Serono, and Roche.

Lancet. Published online July 20, 2010.