Phase III Prospective Randomized Comparison Trial of Depot Octreotide Plus Interferon Alfa-2b Versus Depot Octreotide Plus Bevacizumab in Patients With Advanced Carcinoid Tumors: SWOG S0518

Abstract

Purpose Treatment options for neuroendocrine tumors (NETs) remain limited. This trial assessed the progression-free survival (PFS) of bevacizumab or interferon alfa-2b (IFN-α-2b) added to octreotide among patients with advanced NETs. Patients and Methods Southwest Oncology Group (SWOG) S0518, a phase III study conducted in a US cooperative group system, enrolled patients with advanced grades 1 and 2 NETs with progressive disease or other poor prognostic features. Patients were randomly assigned to treatment with octreotide LAR 20 mg every 21 days with either bevacizumab 15 mg/kg every 21 days or 5 million units of IFN-α-2b three times per week. The primary end point was centrally assessed PFS. This trial is registered with ClinicalTrials.gov as NCT00569127. Results A total of 427 patients was enrolled, of whom 214 were allocated to bevacizumab and 213 to IFN-α-2b. The median PFS by central review was 16.6 months (95% CI, 12.9 to 19.6 months) in the bevacizumab arm and was 15.4 months (95% CI, 9.6 to 18.6 months) in the IFN arm (hazard ratio [HR], 0.93; 95% CI, 0.73 to 1.18; P = .55). By site review, the median PFS times were 15.4 months (95% CI, 12.6 to 17.2 months) for bevacizumab and 10.6 months (95% CI, 8.5 to 14.4 months) for interferon (HR, 0.90; 95% CI, 0.72 to 1.12; P = .33). Time to treatment failure was longer with bevacizumab than with IFN (HR, 0.72; 95% CI, 0.58 to 0.89; P = .003). Confirmed radiologic response rates were 12% (95% CI, 8% to 18%) for bevacizumab and 4% (95% CI, 2% to 8%) for IFN. Common adverse events with bevacizumab and octreotide included hypertension (32%), proteinuria (9%), and fatigue (7%); with IFN and octreotide, they included fatigue (27%), neutropenia (12%), and nausea (6%). Conclusion No significant differences in PFS were observed between the bevacizumab and IFN arms, which suggests that these agents have similar antitumor activity among patients with advanced NETs.

Fig 1.

CONSORT diagram.

Fig 2.

Survival plots by treatment arm: (A) central-review progression-free survival (PFS); (B) site-review (investigator-reviewed) PFS; (C) overall survival; (D) time to treatment failure.

Fig 3.

Waterfall plots: (A) bevacizumab; (B) interferon. The bars on each plot represent the largest decrease under baseline of the sum of longest diameters of all target measurable lesions or, if no decrease was observed, the smallest increase in the sum of longest diameters of target measurable lesions. Patients in whom the smallest increase in measurable lesions was greater than 100% over baseline had data truncated at 100%. Data for patients in whom the best response was progression because of new lesions, death (as a result of disease), or clear worsening of nonmeasurable disease are represented by a bar that shows a 100% increase. Data for patients in whom the best response could not be determined because of symptomatic deterioration or early death (before any follow-up assessments and clearly not as a result of disease) are represented by a gray bar showing 100% increase. Data for patients in whom the best response could not be determined because of inadequate assessment are represented on the far-left side of the plot with a blue bar that shows a 100% increase.

Fig 4.

Survival plots by biomarker levels: (A) central-review progression-free survival (PFS) by chromogranin A (CGA); (B) overall survival by CGA; (C) central-review PFS by neuron-specific enolase (NSE); (D) overall survival (OS) by NSE; (E) central-review PFS by 5-hydroxyindoleacetic acid (5HIAA); and (F) overall survival by 5HIAA. HR, hazard ratio; NR, not reported.