A randomized placebo-controlled phase III study of docetaxel/carboplatin with celecoxib in patients with advanced non-small cell lung cancer (NSCLC): The NVALT-4 study

Reviewer: Geoffrey Geiger MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 1, 2009

Presenter: Groen, M. M. Hochstenbag, J. W. van Putten, A. Vincent, O. Dalesio, B. Biesma, H. J. Smit, A. Termeer, B. E. van den Borne, F. M. Schramel
Presenter's Affiliation: University Medical Center Gron, Netherlands
Type of Session: Scientific


  • Cyclooxygenase-2 (COX-2) is an enzyme responsible for the conversion of arachidonic acid to prostaglandins and is overexpressed in some lung cancers, particularly adenocarcinomas.
  • Increased COX-2 expression has been found to be correlative with VEGF overexpression, increased tumor proliferation and angiogenesis. For example, one study examined 160 patients undergoing resection for stage I disease and COX-2 expression was examined immunohistochemically. Although this study only examined resected stage I cancers, the median survival times for the strong, intermediate or weak expressers were 1.0, 5.5 and 8.5 years, respectively (Achiwa, Clin Can Res 1999). These results have been corroborated with other cancers, including breast and renal cancers.
  • Early investigations examining the additive toxicity of celecoxib did not find overlapping toxicities when employed with full dose carboplatin and taxanes.

Materials and Methods

  • This study is a phase III study involving 540 patients with pathologically proven non-small cell lung cancer (NSCLC) from 32 centers.
  • Exclusion criteria included exposure to prior chemotherapy, NYHA class II-IV heart failure, atherosclerotic diseases, gastrointestinal bleeding, symptomatic brain metastases or chronic use of NSAIDs. However, aspirin (ASA) use of less than 150 mg per day was allowed.
  • Pts were treated with docetaxel 75 mg/m², carboplatin (AUC = 6 mg/ml/min) every 3 wks for 5 cycles and randomized for celecoxib 400 mg daily, starting on day 1 for 3 years or placebo 400 mg daily.
  • Stratification was by WHO performance status (0-1 vs. 2), stage (IIIB vs. IV), ASA (yes versus no) and by treating hospital.
  • The primary endpoint was overall survival.


  • A total of 540 patients were included in the current analysis from a randomization of 561 patients from July 2003 until December of 2007.
  • Of these 561 patients, 281 were randomized to the celecoxib arm and 280 to the placebo arm.
  • Median follow-up was 36 months and the median age was 61 years.
  • Of the randomized patients, 63% were male. 45% of the patients had a performance status (PS) of zero, 48% had a PS of 1 and 6% had a PS of 2.
  • Histologically, 48% were adenocarcinomas, 27% were large cell tumors, 18% were squamous cell cancers and 7% were classified as ‘other’ histologies.
  • In terms of stage, 17% were stage IIIB and the remainder metastatic.
  • Reasons treatment ended included completion of therapy in 51% of the celecoxib arm and 45% in the placebo arm, progression of disease in 17% of the celecoxib arm and 22% in the placebo arm, adverse events in 8% of the celecoxib arm and 10% in the placebo arm, and death in 8% of each arm respectively.
  • Therapy was well tolerated overall with only mild toxicity and no increase in cardiovascular events were observed in the celecoxib arm during the examined period. Adverse events greater than grade 3 with respect to hematological and non-hematological toxicities were also not appreciably different between the treatment arms.
  • Complete response was observed in 0 and 1% and partial response in 33 and 26%, in the celecoxib and placebo arms, respectively. Stable disease was observed in 35 and 40% and progressive disease in 14 and 19%, in the celecoxib and placebo arms, respectively.
  • The partial response rate in evaluable patients was better in the celecoxib arm (p = 0.05). Median progression free survival (PFS) was 5.5 months and overall survival (OS) was 8.3 months but was not statistically significant when compared between the treatment groups.
  • Hazard ratios stratified by ASA and PS for progression free survival and overall survival were 0.94 (95% CI=0.79-1.13) and 0.95 (0.79-1.15), respectively.

Author's Conclusions

  • The addition of celecoxib to docetaxel and carboplatin in patients with stage IIIB or IV lung cancers improves response rate but not the progression free interval or overall survival.
  • No difference was observed in the progression free interval or overall survival when examining outcomes by histology although there was a trend towards better outcomes in squamous cell histologies.
  • No additional cardiovascular toxicity was observed in patients receiving celecoxib as compared to the placebo arm.

Clinical/Scientific Implications

  • This study examined the hypothesis that celecoxib administration in conjunction with standard first line therapy for stage IIIB or IV NSCLC would improve outcomes through inhibition of COX-2 overexpression and may improve the response to chemotherapy.
  • Based on the results of this study, celecoxib administration cannot be routinely advocated for patients with stage IIIB or IV NSCLC, although there is not additive hematological, non-hematological or cardiovascular toxicities in patients receiving celecoxib.
  • Ultimately, COX-2 inhibition via celecoxib may improve outcomes in selected patients and demonstrates the importance of obtaining tissue specimens for correlative studies looking at whether celecoxib inhibition improves outcomes in patients whose tumors are found to overexpress COX-2 immunohistochemically. For example, CALGB 30203 examined eicosanoid modulation in advanced non-small cell lung cancer and found that COX-2 overexpression is a negative prognostic factor for survival but a positive predictive factor if patients received celecoxib.
  • CALGB 30801 is an upcoming study testing the hypothesis that COX-2 inhibition in patients overexpressing COX-2 improves outcomes and will help answer the question of whether celecoxib would be beneficial in certain subpopulations of patients.