Final results of the phase III, randomized, double-blind AVOREN trial of first-line bevacizumab (BEV) + interferon-?2a (IFN) in metastatic renal cell carcinoma (mRCC)

Reviewer: Geoffrey Geiger MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 3, 2009

Presenter: B.J. Escudier, et al.
Presenter's Affiliation: Institut Gustave Roussy, Villejuif, France
Type of Session: Scientific


  • Renal cell carcinoma (RCC) is diagnosed in more than 120,000 patients in the United States and Europe every year.
  • The most common histology is clear cell carcinoma.
  • The 5-year survival rate for patients with stage IV metastatic RCC (mRCC) is 10-20% and a third of patients present with metastatic disease at the time of diagnosis.
  • Metastatic RCC is highly resistant to conventional treatments and until recently, the standard of care for these patients was immunotherapy with either interleukin 2 or interferon, both of which have only modest efficacy and response rates (<20%) but confer substantial toxicities.
  • Trials have demonstrated that treatment with interferon results in a median overall survival of 13 months and high dose interleukin 2 can achieve curative outcomes in a small subset of patients, estimated at 5-10%
  • Tyrosine kinase inhibitors (e.g., sorafenib, sunitinib) are also used in the treatment of advanced RCC in patients that have failed interferon or interleukin 2.
  • Bevacizumab is a humanized monoclonal antibody against the VEGF receptor, which inhibits angiogenesis, a rate-limiting step for cancer growth that has demonstrated clinical benefit in phase II studies of mRCC.
  • In the original reporting of the AVOREN trial, bevacizumab plus interferon was demonstrated to significantly improve progression free survival as compared to placebo plus interferon in patients with mRCC (Escudier, Lancet. 2007).
  • The study was unblinded after interim analysis demonstrated significant superiority of bevacizumab plus interferon and the trial committee recommended crossover of patients from placebo to bevacizumab.
  • This report summarizes the final results of the AVOREN trial.

Materials and Methods

  • Trial design: international, multicenter, randomized, double blind phase III trial with patients randomized in 1:1 fashion to receive bevacizumab plus interferon alfa or placebo plus interferon alfa. Randomization was done centrally and patients stratified according to country and Memorial Sloan Kettering Cancer Center (MSKCC) risk group (favorable, intermediate, or poor).
  • Patients were eligible for enrollment if they were 18 years or older, with predominately clear cell histology (>50%) and had undergone nephrectomy or partial nephrectomy, Karnofsky performance score, and had normal hepatic, hematopoietic, and renal function.
  • Exclusion criteria included prior systemic therapy for mRCC, recent major surgical procedures, evidence of brain metastases, ongoing full-dose oral or parenteral anticoagulant or antiplatelet therapy, uncontrolled hypertension on medication, clinically significant cardiac disease or chronic corticosteroid use.
  • Patients were randomized to interferon (9 MIU three times weekly as a subcutaneous injection for a maximum of 52 weeks) plus bevacizumab (10 mg/kg given IV every 2 weeks) or placebo until disease progression, unacceptable toxicity or withdrawal of consent was given.
  • The primary endpoint was overall survival (OS) with pre-planned stratified and unstratified analyses according to regional regulatory requirements. Secondary endpoints included progression-free survival (PFS), overall response rates and safety.
  • Tumor measurements and assessments were performed with imaging studies performed every 8 weeks up to week 32 and every 12 weeks thereafter until evidence of disease progression.


  • 649 pts were randomized to bevacizumab plus interferon or (n = 327) or placebo plus interferon (n = 322).
  • The separate arms of the trial were well balanced with respect to sex, age, baseline VEGF expression, presence or absence of lung metastases, number of metastatic sites and MSKCC score.
  • An independent radiological review confirmed previous investigator assessments, demonstrating a PFS of 10.4 vs. 5.5 months (HR 0.57) and a response rate of 31% vs. 12% in the two arms, verifying the robustness of the investigator analysis.
  • At the time of final OS analysis (444 events), median follow-up was 22.9 months in the bevacizumab arm and 20.6 months in the placebo arm.
  • The final median OS stratified for treatment region and Motzer score was 23.3 months in the bevacizumab plus interferon arm and 21.3 months in the interferon plus placebo arm with a HR 0.86 [95% CI: 0.72-1.04] and p value of 0.1291.
  • No new or unexpected adverse events were observed. More patients in the interferon plus placebo arm than the bevacizumab plus interferon arm received post-protocol therapy, including therapy with tyrosine kinase inhibitors, mTOR inhibitors, cytokine therapy and chemotherapy: a total of 180 (55%) patients in the bevacizumab plus interferon arm and 202 (63%) patients in the interferon plus placebo arm.
  • Exploratory analysis showed that median OS in patients receiving second-line tyrosine kinase inhibitors (BEV + IFN, n = 96; IFN + placebo, n = 81) was 38.6 months vs. 33.2 months (HR = 0.77 [95% CI: 0.51-1.15], p = 0.1948).
  • In a multiple Cox regression model, further adjustment on subgroup analysis (accounting for sex, age, baseline VEGF, presence or absence of lung metastases, number of metastatic sites and MSKCC score) improved the treatment effect with an HR of 0.78 and p = 0.0219.

Author's Conclusions

  • Bevacizumab plus interferon represents a first-line standard of care for pts with mRCC, improving PFS and objective response rate (ORR), with a trend towards improved overall survival compared to interferon.
  • Bevacizumab plus interferon treatment was well tolerated with respect to the interferon plus placebo arm with little increase in reported adverse events.
  • Further adjustment in multiple Cox regression model improved the treatment effect with an HR of 0.78 and p = 0.0219.
  • The observed OS results may have been influenced by subsequent anti-neoplastic therapy, which was not prespecified in the protocol and represents an uncontrolled element of the OS analysis.

Clinical/Scientific Implications

  • Overall, this studyÕs updated results were disappointing with an overall survival that was not statistically significant with a HR of 0.86 and p = 0.1291. However, PFS and ORR favored the bevacizumab plus interferon arm.
  • Overall, the treatment was well tolerated and there was a trend towards improved OS with bevacizumab plus interferon vs. interferon plus placebo.
  • Overall, the results are very similar to CALGB 90206, a similar but superiorly powered study with slightly lower overall survival numbers than AVOREN.
  • Some subgroups do very well and have a statistically significant improvement in OS; for example, favorable risk patients without liver metastases that received second line sunitinib (HR of 0.78 and p = 0.0219). A significant mix of patient prognostic factors likely confounds the results
  • Bottom line: Results do not exceed high watermark of sunitinib with an OS of 26.4 months. Therefore, although bevacizumab and interferon represent a reasonable approach to the treatment of mRCC, there is no compelling reason to select this regimen over interferon plus sunitinib, where at ASCO 2008, the objective response rate was reported as 47% in the sunitinib arm versus 12% in the interferon group (P<0.000001) and median overall survival was 26.4 months for sunitinib-treated patients versus 21.8 months (P=0.051). Contrast this to the results of CALGB 90206 and AVOREN, whose overall survival numbers range from (20-23 months). (Figlin, JCO 26: 2008 Abstract # 5024).

See Patient Summary