A Randomized Phase III Trial Comparing Bexarotene/Carboplatin/Paclitaxel vs. Carboplatin/Paclitaxel in Chemotherapy-Naive Patients with Advanced or Metastatic NSCLC (SPIRIT II Trial)

Reviewer: S. Jack Wei, MD
University of Pennsylvania School of Medicine
Last Modified: May 15, 2005

Presenter: G.R. Blumenschein, Jr., MD
Presenter's Affiliation: SPIRIT II Trial Group
Type of Session: Scientific


  • High RXR-beta expression has been shown to be correlated with increased survival in patients with non-small cell lung cancer (NSCLC).
  • Bexarotene (Targretin) is an RXR-selective rexinoid which forms homodimers and heterodimers with a number of molecular partners that ultimately lead to increased apoptosis and alters cell cycle control, differentiation, anti-metastatic activity, and anti-angiogenic activity.
  • The use of bexarotene in patients with NSCLC has been shown to lead to disease stabilization and increased survival in several phase I and phase II trials.
  • Bexarotene has also been found to result in increased triglyceride (TGC) levels and decreased thyroid stimulating hormone levels in a number of patients.
  • This trial, in conjunction with the SPIRIT I Trial, was designed to determine if the addition of bexarotene to standard chemotherapy improved outcomes in patients with advanced or metastatic NSCLC.

Materials and Methods

  • Patients with NSCLC who were stage IIIB with pleural effusion or stage IV were randomized to:
              1)      Carboplatin (AUC=6) + paclitaxel (200mg/m2) q3wks
              2)      Carboplatin (AUC=6) + paclitaxel (200mg/m2) q3wks + bexarotene (400 mg/m2 qd)
  • Eligible patients were ECOG performance status 0 or 1 and had not received a prior chemotherapy.
  • Patients receiving bexarotene were allowed to continue bexarotene until disease progression after completion of chemotherapy.
  • Anti-lipid therapy was started on day one for patients receiving bexarotene.
  • Primary endpoint was overall survival (OS) with secondary endpoint of 2-year survival.
  • Additional endpoints included progression-free survival (PFS), time to progression (TTP), safety, population pharmacokinetics, and biochemical and genetic biomarkers.


  • A total of 612 patients were enrolled.
  • Treatment groups were well-balanced with respect to gender, stage, performance status, age, histopathology, ethnicity, smoking history, and degree of weight loss.
  • Median survival was no different between the control group and the bexarotene group (9.2 mo vs. 8.5 mo, p=0.2).
  • 2-year survival was no different between the two groups (16.3% vs. 12.4%, p=0.2).
  • There was no difference in median PFS (4.9 mo vs. 4.1 mo, p=0.06).
  • There were significant increases with the use of bexarotene with respect to grade 3/4 neutropenia, asthenia, leukopenia, and dehydration.
  • 32% of patients receiving bexarotene experienced grade 3/4 hypertriglyceridemia compared to 0% in the control arm.
  • Retrospective subgroup analysis showed that patients hypertriglyceridemia correlated with improved outcome with the use of bexarotene.
  • It was hypothesized that hypertriglyceridemia served as a surrogate biomarker for response to treatment and this was analyzed further.
  • Patients with NCI grade 3 or 4 hypertriglyceridemia had improved survival with bexarotene with a hazard ratio of 0.82 (p=0.0002).
  • When separated by triglyceride level, patients with NCI grade 3 or 4 triglycerides had a median survival of 12.4 mo vs. 9.2 mo for the control group (p=0.014)
  • Patients with NCI grade 0-2 triglycerides had a median survival of 6.6 mo vs. 9.2 mo for the control group (p=0.001).
  • The correlation of high triglycerides with improved outcomes was found for all patient subgroups.

Author's Conclusions

  • Bexarotene added to carboplatin/paclitaxel in the first-line setting did not improve overall survival.
  • Retrospective subset analysis suggests that elevated triglycerides in patients treated with  bexarotene correlates with survival.
  • The relationship between hypertriglyceridemia and response to bexarotene warrants further study.
  • Additional biochemical and genetic marker analyses could help identify likely responders to treatment with bexarotene.

Clinical/Scientific Implications

The results presented in this study showed no benefit to the overall population with the addition of bexarotene to standard chemotherapy with carboplatin and paclitaxel in the first-line setting with advanced and metastatic NSCLC.  These results are disappointing given the promising results seen in earlier phase studies with this agent.  There was no indication that bexarotene benefited the overall population, and in fact, there was a trend towards worsened PFS with bexarotene.  These results mirrored the results of the SPIRIT I trial which was also presented at this meeting.  In that trial, bexarotene was given with cisplatin and vinorelbine in a similar population of patients, and again, there was no improvement in median survival, OS, or PFS.
Interestingly, in both trials, patients with elevated triglycerides did show improvement in survival.  This was somewhat paradoxical, because patients who experienced high elevations in triglycerides despite antilipid therapy had the doses of bexarotene reduced in both trials.  Therefore, the median dose of bexarotene received in patients with elevated bexarotene was significantly lower than the dose received by patients who did not have triglyceride elevation.  The exact cause of the relationship between triglyceride level and response to bexarotene is unclear.  It is possible that increased triglycerides serve as a biomarker for response or inherent sensitivity to the experimental drug.  This relationship warrants further investigation.  If it is possible to identify patients who will experience this elevation of triglycerides in response to treatment prior to the initiation of therapy, a susceptible subset of these patients could be targeted for treatment with bexarotene.  Until then, the use of bexarotene should be confined to experimental protocols.