Bevacizumab, Erlotinib, Not Recommended in Combined-Modality Stage III NSCLC Setting
In addition to “lack of an efficacy signal” with respect to progression-free survival (PFS) or overall survival (OS), there is a “substantial risk of esophageal toxicity,” noted lead author Mark A. Socinski, MD, of the UPMC Lung Cancer Center of Excellence, Pittsburgh, PA.
Previously, both bevacizumab and erlotinib have been shown to improve survival in stage IV NSCLC and, when the trial was designed, it was believed these agents could improve locoregional or systemic control—or both.
“To our knowledge, this trial is the only trial that evaluated bevacizumab in the combined-modality setting in lung cancer and completed accrual,” he stated. “However, the PFS and OS results observed on this trial were unimpressive compared with our previous experiences.”
The study enrolled 45 patients. Induction chemotherapy with carboplatin AUC 6, paclitaxel 225mg/m2, and bevacizumab 15 mg/kg was administered on days 1 and 22 followed by concurrent chemotherapy with carboplatin AUC 2 and paclitaxel 45 mg/m2 weekly with bevacizumab 10 mg/kg every other week for 4 doses and thoracic conformal radiation therapy to 74 Gy.
Median age of the patients was 61 years (range, 34 to 74) and approximately half were men; 64% had stage IIIA disease, 60% of which was adenocarcinoma. The majority (71%) had ECOG PS of 0.
In phase 1, the first cohort received no erlotinib; the second, erlotinib 100mg; and the third, erlotinib 150mg Tuesday through Friday during thoracic conformal radiation therapy. Consolidation therapy with erlotinib 150mg/day and bevacizumab 15 mg/kg every 3 weeks was planned 3 to 6 weeks later for 6 cycles.
Objective response rates to induction were 39% (95% CI, 24%–55%) and, to overall treatment, 60% (95% CI, 44%–75%). Median PFS was 10.2 months (95% CI, 8.4–18.3 months) and OS, 18.4 months (95% CI, 13.4–31.7 months).
Esophagitis was the principal toxicity: 29% was grade 3 or 4 and one patient had grade 3 tracheoesophageal fistula that was often prolonged.
An accompanying editorial noted, “The take-home message is that in retrospect, understanding how difficult it is to design and implement clinical trials in the current era of rapid progress in molecular understandings of NSCLC, this particular trial might have been designed differently.”