Randomized phase III trial of capecitabine/cisplatin (XP) vs. continuous infusion of 5FU/cisplatin (FP) as first line therapy in patients (pts) with advanced gastric cancer (AGC): Efficacy and safety results.

Reviewer: Christopher Dolinsky, MD
University of Pennsylvania School of Medicine
Last Modified: June 5, 2006

Presenter: Y. Kang
Presenter's Affiliation: Asan Medical Center, Seoul, Korea
Type of Session: Scientific


  • Capecitabine is an oral fluropyrimidine with proven efficacy and favorable safety in metastatic colorectal cancer.
  • Because it is given orally, patients who receive capecitabine are spared a central line and hospitalization time.
  • Continuous infusion 5FU plus bolus i.v. cisplatin has been a standard/reference regimen in advanced gastric cancer.
  • Phase II data support the testing of capecitabine/cisplatin (XP) in advanced gastric cancer (55% response rate (RR), median time to progression 5.8 months).

Materials and Methods

  • A multi-national prospective phase III trial randomized patients with advanced gastric cancers to either bolus i.v. cisplatin (80mg/m2 d1) with capecitabine (1000mg/m2 bid d1-14) given q3weeks, or bolus i.v. cisplatin (80mg/m2 d1) with continuous infusion 5FU (800 mg/m2 d1-5) given q3weeks.
  • Patients were treated until disease progression or unacceptable toxicity.
  • The primary endpoint was non-inferiority of XP in terms of progression free survival.
  • Secondary endpoints included response rate, duration of response, time to progression, overall survival and safety.
  • Patients were ages 18 to 75 and had to have a Karnofsky score >70%.
  • All patients had either advanced or metastatic gastric cancer with at least 1 measurable lesion and could not have had any prior therapy for their cancer.
  • 316 patients were enrolled in 42 centers across 13 countries.
  • The treatment arms were well balanced with regards to patient and disease related factors.


  • Compared to the FP patients, the XP patients had a trend towards improved progression free survival (HR 0.81, 95%CI 0.63-1.03, p=0.08).
  • This improvement in PFS remained robust amongst all subgroups analyzed.
  • The two arms had similar overall survival: XP median OS 10.5 months (95%CI 9.3-11.2 months) and FP median OS 9.3 months (95%CI 7.4-10.6 months).
  • The XP patients had improved response rates compared to the FP patients (RR 41% vs 29%, p=0.03).
  • Both regimens were well tolerated, with small numbers of adverse events in each arm.
  • The XP patients had higher rate of hand-foot syndrome (5% vs 0%), and the FP patients had a higher rate of stomatitis (6% vs. 2%); otherwise there were no differences seen in the safety profiles of these regimens.
  • Treatment withdrawal and mortality were similar in both arms.

Author's Conclusions

  • The primary objective was met in that XP was not proven to be inferior to FP in terms of progression free survival.
  • Response rates were superior in the XP arm.
  • Multivariate and subgroup analyses confirm the results.
  • There is a similar and favorable safety profile of XP and FP.
  • XP avoids the inconvenience and complications associated with infusional FP.

Clinical/Scientific Implications

The authors presented the results of well designed multi center randomized phase III study comparing XP to FP in advanced gastric cancer.  The authors’ conclusions and methods were valid, and they have demonstrated the non-inferiority of XP compared to FP.  Capecitabine is a relatively new chemotherapeutic agent that is quite expensive.  It would be interesting for someone to perform an economic analysis of these two regimens.  There are certainly benefits to giving a medication orally over intravenously, however, there is one particular drawback.  When patients are treated intravenously, their physicians have complete control over the delivery of the prescribed medication.  When medications are given orally, then patients become in charge of their own care.  Studies have shown that the majority of patients prescribed antibiotics twice a day do not take the entire course as directed.  It would be interesting to try and measure patients’ compliance with their capecitabine prescription in the next clinical trial that employs it.