Skip to content
Bending the Curve Trial by Trial
?

RATIONALE-309

Tislelizumab plus chemotherapy as first-line treatment for recurrent or metastatic nasopharyngeal cancer

Phase 3N = 263NCT03924986

Population
Treatment-naive recurrent or metastatic nasopharyngeal cancer
Design
Double-blind, placebo-controlled phase 3 trial conducted in Asia. 263 treatment-naive patients with recurrent or metastatic NPC were randomised 1:1 to tislelizumab 200 mg or placebo every 3 weeks, both with gemcitabine and cisplatin for 4 to 6 cycles; placebo-arm participants could cross over to tislelizumab monotherapy at progression. The primary end point was progression-free survival assessed by an independent review committee.
Primary endpoint
PFS
Arms
  • Tislelizumab plus gemcitabine-cisplatin (experimental, n = 131): Tislelizumab 200 mg every 3 weeks with gemcitabine and cisplatin for 4 to 6 cycles
  • Placebo plus gemcitabine-cisplatin (control, n = 132): Placebo every 3 weeks with gemcitabine and cisplatin for 4 to 6 cycles
People
Yang Y, senior author Fang W
PFSHR 0.52 (95% CI 0.38–0.73)PFSHR 0.53 (95% CI 0.39–0.71)Δ median PFS+2.2 monthsOSHR 0.73 (95% CI 0.51–1.05)Δ median OS+13.5 months
Show the numbers
Results reported for RATIONALE-309
EndpointAnalysisTislelizumab plus gemcitabine-cisplatinPlacebo plus gemcitabine-cisplatinEffect
PFSRandomised patients (N = 263)interimYang 2023——HR 0.52 (95% CI 0.38–0.73)P<0.0001Medians are not given in the abstract. PFS benefit was observed regardless of PD-L1 expression.
PFSRandomised patients (N = 263)long term-updatemedian follow-up 27.5 monthsYang 2026median 9.6 months (7.6–11.6)median 7.4 months (5.6–7.6)HR 0.53 (95% CI 0.39–0.71)Secondary analysis after about 3 years; the abstract gives median follow-up as 27.5 months (range 0.1-53.0).
OSRandomised patients (N = 263)long term-updatemedian follow-up 27.5 monthsYang 2026median 45.3 monthsmedian 31.8 monthsHR 0.73 (95% CI 0.51–1.05)Placebo-arm participants could cross over to tislelizumab monotherapy at progression. Rank-preserving structural failure time analysis: HR 0.56 (95% CI 0.27-1.19); two-stage crossover-adjusted analysis: HR 0.62 (95% CI 0.40-0.97). The 95% CIs for the medians (33.4 to not estimable; 25.0 to not estimable) have an upper bound that cannot be entered as a number.
Results reported for RATIONALE-309
EndpointAnalysisTislelizumab plus gemcitabine-cisplatinPlacebo plus gemcitabine-cisplatinEffect
PFSRandomised patients (N = 263)interimYang 2023——HR 0.52 (95% CI 0.38–0.73)P<0.0001Medians are not given in the abstract. PFS benefit was observed regardless of PD-L1 expression.
PFSRandomised patients (N = 263)long term-updatemedian follow-up 27.5 monthsYang 2026median 9.6 months (7.6–11.6)median 7.4 months (5.6–7.6)HR 0.53 (95% CI 0.39–0.71)Secondary analysis after about 3 years; the abstract gives median follow-up as 27.5 months (range 0.1-53.0).
OSRandomised patients (N = 263)long term-updatemedian follow-up 27.5 monthsYang 2026median 45.3 monthsmedian 31.8 monthsHR 0.73 (95% CI 0.51–1.05)Placebo-arm participants could cross over to tislelizumab monotherapy at progression. Rank-preserving structural failure time analysis: HR 0.56 (95% CI 0.27-1.19); two-stage crossover-adjusted analysis: HR 0.62 (95% CI 0.40-0.97). The 95% CIs for the medians (33.4 to not estimable; 25.0 to not estimable) have an upper bound that cannot be entered as a number.
Hazard ratio (95% CI) by endpoint. Values below 1 favour the experimental arm.
PFSPFSOS0.512◀ experimentalcontrol ▶
Data table
EndpointHR95% CI
PFS0.520.38–0.73
PFS0.530.39–0.71
OS0.730.51–1.05
Illustrative — modelled from reported values, not trial data. PFS, long term-update.
0%25%50%75%100%05101520Months (modelled)
  • Tislelizumab plus gemcitabine-cisplatin (experimental)
  • Placebo plus gemcitabine-cisplatin (control)

Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.

Data table (modelled values)
MonthsTislelizumab plus gemcitabine-cisplatinPlacebo plus gemcitabine-cisplatin
0100.0%100.0%
664.8%57.0%
1242.0%32.5%
1827.3%18.5%
2417.7%10.6%
Illustrative — modelled from reported values, not trial data. OS, long term-update.
0%25%50%75%100%020406080100Months (modelled)
  • Tislelizumab plus gemcitabine-cisplatin (experimental)
  • Placebo plus gemcitabine-cisplatin (control)

Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.

Data table (modelled values)
MonthsTislelizumab plus gemcitabine-cisplatinPlacebo plus gemcitabine-cisplatin
0100.0%100.0%
2865.2%54.3%
5741.8%28.9%
8527.2%15.7%
11317.7%8.5%
Primary publication: Yang Y, Pan J, Wang H, Zhao Y, Qu S, Chen N, et al. Tislelizumab plus chemotherapy as first-line treatment for recurrent or metastatic nasopharyngeal cancer: A multicenter phase 3 trial (RATIONALE-309). Cancer Cell. 2023;41(6):1061-1072.e4. PubMedDOI

Appears in

Sources

  1. Yang Y, Pan J, Wang H, Zhao Y, Qu S, Chen N, et al. Tislelizumab plus chemotherapy as first-line treatment for recurrent or metastatic nasopharyngeal cancer: A multicenter phase 3 trial (RATIONALE-309). Cancer Cell. 2023;41(6):1061-1072.e4.

    PubMedDOI

  2. Yang Y, Yen CJ, Pan J, Wang H, Qu S, Chen N, et al. First-Line Tislelizumab Plus Chemotherapy for Recurrent or Metastatic Nasopharyngeal Cancer: Three-Year Follow-Up of the Phase 3 RATIONALE-309 Randomized Clinical Trial. JAMA Oncol. 2026;12(4):384-393.

    PubMedDOI