ClarIDHy: trial | Bending the Curve: Trial by Trial

ClarIDHy

Needs review

A Phase 3, Multicenter, Randomized, Double-Blind, Placebo-controlled Study of AG-120 in Previously-treated Subjects With Nonresectable or Metastatic Cholangiocarcinoma With an IDH1 Mutation (ClarIDHy)

Phase 3N = 187NCT02989857

Population

Adults with histologically confirmed, advanced IDH1-mutant cholangiocarcinoma that had progressed on previous therapy (up to two previous regimens for advanced disease), ECOG performance status 0 or 1, and a measurable lesion (RECIST 1.1).

Design

Multicentre, randomised (2:1), double-blind, placebo-controlled phase 3 study at 49 hospitals in six countries: ivosidenib 500 mg or matched placebo once daily in continuous 28-day cycles, stratified by number of previous systemic regimens. Crossover from placebo to ivosidenib was permitted on radiological progression. Primary endpoint: progression-free survival by independent central review. The primary analysis (data cutoff 31 January 2019) included 185 patients (124 vs 61); the final overall survival analysis included 187 patients (126 vs 61; 43 placebo patients crossed over).

Primary endpoint

PFS

Arms

Placebo (control, n = 61): Matched placebo orally once daily; crossover to ivosidenib permitted on radiological progression
Ivosidenib (experimental, n = 126): Ivosidenib 500 mg orally once daily in continuous 28-day cycles

People

Abou-Alfa GK

PFSHR 0.37 (95% CI 0.25–0.54)Δ median PFS+1.3 monthsOSHR 0.79 (95% CI 0.56–1.12)Δ median OS+2.8 monthsOSHR 0.49 (95% CI 0.25–1.24)OSHR 0.49 (95% CI 0.34–0.70)

Show the numbers

Results reported for ClarIDHy
Endpoint	Analysis	Ivosidenib	Placebo	Effect
PFSITT at the primary analysis (185: ivosidenib 124, placebo 61)Progression-free survival by independent central review (intention-to-treat)	primarymedian follow-up 6.9 monthsAbou-Alfa 2020	median 2.7 months (1.6–4.2)	median 1.4 months (1.4–1.6)	HR 0.37 (95% CI 0.25–0.54)P<0.0001 (one-sided)Median follow-up for progression-free survival (interquartile range 2.8 to 10.9 months). The paper has two published corrections (Lancet Oncol 2020;21:e462 and 2024;25:e61) whose content could not be checked against these figures.
OSITT: all randomised patients (187: ivosidenib 126, placebo 61)	finalZhu 2021	median 10.3 months (7.8–12.4)	median 7.5 months (4.8–11.1)	HR 0.79 (95% CI 0.56–1.12)P=.09 (one-sided)Overall survival was the first key secondary endpoint; 43 of the 61 placebo patients crossed over to ivosidenib. Data cutoff 31 May 2020 is from the full text.
OSITT; placebo arm adjusted for crossover with a rank-preserving structural failure time (RPSFT) model; confidence intervals bootstrapped	finalZhu 2021	—	median 5.1 months (3.5–11.4)	HR 0.49 (95% CI 0.25–1.24)Values from the JAMA Oncology erratum of 15 October 2026 (PMID 42879142), which gives these bootstrapped 95% CIs as the corrected intervals for the prespecified RPSFT crossover adjustment in the final overall survival analysis; median OS with placebo remained 5.1 months and the hazard ratio 0.49. The erratum gives no P value for this analysis. The 2021 paper reported the same point estimates with narrower intervals (next result).
OSITT; placebo arm adjusted for crossover with a rank-preserving structural failure time (RPSFT) model; confidence intervals without bootstrapping	exploratoryZhu 2021	—	median 5.1 months (3.8–7.6)	HR 0.49 (95% CI 0.34–0.70)P<.001 (one-sided)As published in the 2021 abstract and Results, where it is the result of the prespecified RPSFT crossover adjustment. The 2026 erratum keeps these values as an additional ad hoc sensitivity analysis without bootstrapping and gives the bootstrapped intervals (previous result) as the corrected ones.

Results reported for ClarIDHy
Endpoint	Analysis	Ivosidenib	Placebo	Effect
PFSITT at the primary analysis (185: ivosidenib 124, placebo 61)Progression-free survival by independent central review (intention-to-treat)	primarymedian follow-up 6.9 monthsAbou-Alfa 2020	median 2.7 months (1.6–4.2)	median 1.4 months (1.4–1.6)	HR 0.37 (95% CI 0.25–0.54)P<0.0001 (one-sided)Median follow-up for progression-free survival (interquartile range 2.8 to 10.9 months). The paper has two published corrections (Lancet Oncol 2020;21:e462 and 2024;25:e61) whose content could not be checked against these figures.
OSITT: all randomised patients (187: ivosidenib 126, placebo 61)	finalZhu 2021	median 10.3 months (7.8–12.4)	median 7.5 months (4.8–11.1)	HR 0.79 (95% CI 0.56–1.12)P=.09 (one-sided)Overall survival was the first key secondary endpoint; 43 of the 61 placebo patients crossed over to ivosidenib. Data cutoff 31 May 2020 is from the full text.
OSITT; placebo arm adjusted for crossover with a rank-preserving structural failure time (RPSFT) model; confidence intervals bootstrapped	finalZhu 2021	—	median 5.1 months (3.5–11.4)	HR 0.49 (95% CI 0.25–1.24)Values from the JAMA Oncology erratum of 15 October 2026 (PMID 42879142), which gives these bootstrapped 95% CIs as the corrected intervals for the prespecified RPSFT crossover adjustment in the final overall survival analysis; median OS with placebo remained 5.1 months and the hazard ratio 0.49. The erratum gives no P value for this analysis. The 2021 paper reported the same point estimates with narrower intervals (next result).
OSITT; placebo arm adjusted for crossover with a rank-preserving structural failure time (RPSFT) model; confidence intervals without bootstrapping	exploratoryZhu 2021	—	median 5.1 months (3.8–7.6)	HR 0.49 (95% CI 0.34–0.70)P<.001 (one-sided)As published in the 2021 abstract and Results, where it is the result of the prespecified RPSFT crossover adjustment. The 2026 erratum keeps these values as an additional ad hoc sensitivity analysis without bootstrapping and gives the bootstrapped intervals (previous result) as the corrected ones.

Hazard ratio (95% CI) by endpoint. Values below 1 favour the experimental arm.

Data table

Endpoint	HR	95% CI
PFS	0.37	0.25–0.54
OS	0.79	0.56–1.12
OS	0.49	0.25–1.24
OS	0.49	0.34–0.7

Illustrative — modelled from reported values, not trial data. PFS, primary.

Show curves

Ivosidenib (experimental)
Placebo (control)

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

Data table (modelled values)

Months	Ivosidenib	Placebo
0	100.0%	100.0%
2	59.8%	37.1%
3	46.3%	22.6%
5	27.7%	8.4%
7	16.6%	3.1%

Illustrative — modelled from reported values, not trial data. OS, final.

Show curves

Ivosidenib (experimental)
Placebo (control)

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

Data table (modelled values)

Months	Ivosidenib	Placebo
0	100.0%	100.0%
6	66.8%	57.4%
13	41.7%	30.1%
19	27.8%	17.3%
26	17.4%	9.0%

Appears in

2020 ClarIDHy: ivosidenib versus placebo in IDH1-mutant cholangiocarcinoma

Sources

Abou-Alfa GK, Macarulla T, Javle MM, Kelley RK, Lubner SJ, Adeva J, et al. Ivosidenib in IDH1-mutant, chemotherapy-refractory cholangiocarcinoma (ClarIDHy): a multicentre, randomised, double-blind, placebo-controlled, phase 3 study. Lancet Oncol. 2020;21(6):796-807.
PubMed DOI
Zhu AX, Macarulla T, Javle MM, Kelley RK, Lubner SJ, Adeva J, et al. Final Overall Survival Efficacy Results of Ivosidenib for Patients With Advanced Cholangiocarcinoma With IDH1 Mutation: The Phase 3 Randomized Clinical ClarIDHy Trial. JAMA Oncol. 2021;7(11):1669-1677.
PubMed DOI