Skip to content
Bending the Curve Trial by Trial
?

Perioperative pembrolizumab for locally advanced thymic epithelial tumors: a single-arm, phase 2 trial

Phase 2N = 40NCT03858582

Population
40 untreated patients with potentially resectable Masaoka-Koga stage III-IV thymic epithelial tumours: WHO B3 thymoma 7 (17.5%), thymic carcinoma 29 (72.5%); stage IV 31 (77.5%)
Design
Single-arm, prospective phase 2 trial (Samsung Medical Center, Seoul, per the author affiliations; the abstract's registration line reads NCT03858582 (MK3475-971)) in patients with potentially resectable Masaoka-Koga stage III-IV thymic epithelial tumours: neoadjuvant docetaxel, cisplatin and pembrolizumab every 3 weeks for 3 cycles, then surgery, then maintenance pembrolizumab for 2 years; patients with R1 or R2 resection received adjuvant radiotherapy with pembrolizumab. 40 untreated patients were recruited between March 2020 and January 2025; 28 (70.0%) underwent surgical resection. The primary end point was major pathologic response (MPR). Median follow-up 27.5 months. Two patients with thymoma died from myocarditis. ClinicalTrials.gov (NCT03858582) describes the stage as modified Masaoka III or IVa, lists a start of 29 April 2019, status unknown (last verified March 2021) and an estimated enrolment of 40; the paper reports recruitment from March 2020 to January 2025 and stage III-IV. Both are shown as published.
Primary endpoint
Major pathological response
Arms
  • Docetaxel, cisplatin and pembrolizumab, surgery, then pembrolizumab (experimental, n = 40): Neoadjuvant docetaxel 75 mg/m2, cisplatin 75 mg/m2 and pembrolizumab 200 mg every 3 weeks for 3 cycles; surgery; maintenance pembrolizumab for 2 years; adjuvant radiotherapy with pembrolizumab after R1 or R2 resection
People
Bang YH, senior author Park S
Show the numbers
Results reported for Perioperative pembrolizumab for locally advanced thymic epithelial tumors: a single-arm, phase 2 trial
EndpointAnalysisDocetaxel, cisplatin and pembrolizumab, surgery, then pembrolizumabControlEffect
Major pathological response28 patients who underwent surgical resection (13 with MPR)primarymedian follow-up 27.5 monthsBang 202546.4%—The abstract states that all cases of MPR occurred in patients with thymic carcinoma and none in thymoma. The denominator is the 28 resected patients, not all 40 recruited.
pCR28 patients who underwent surgical resection (5 with pathologic complete response)finalBang 202517.9%—All pathologic complete responses occurred in thymic carcinoma, none in thymoma (abstract).
DFSPatients who underwent surgical resection (denominator not restated in the abstract)Disease-free survival measured from the time of surgeryfinalmedian follow-up 27.5 monthsBang 2025median 49.3 months; 91% at 12 mo (79.9–100.0%)—Median DFS 95% confidence interval 25.3 months to not reached (upper bound not estimable). Median follow-up for the whole group was 27.5 months (95% CI 22.0-39.2).
Grade 3 adverse eventsAll 40 patients (9 with grade 3 adverse events as printed)finalBang 202522.5%—The abstract does not say whether counts are patients or events, or whether all adverse events or only treatment-related ones are counted.
Grade 4 adverse eventsAll 40 patients (5 with grade 4 adverse events as printed)finalBang 202512.5%—Same caveat as for grade 3 events. The abstract reports that two patients with thymoma died from myocarditis.
Results reported for Perioperative pembrolizumab for locally advanced thymic epithelial tumors: a single-arm, phase 2 trial
EndpointAnalysisDocetaxel, cisplatin and pembrolizumab, surgery, then pembrolizumabControlEffect
Major pathological response28 patients who underwent surgical resection (13 with MPR)primarymedian follow-up 27.5 monthsBang 202546.4%—The abstract states that all cases of MPR occurred in patients with thymic carcinoma and none in thymoma. The denominator is the 28 resected patients, not all 40 recruited.
pCR28 patients who underwent surgical resection (5 with pathologic complete response)finalBang 202517.9%—All pathologic complete responses occurred in thymic carcinoma, none in thymoma (abstract).
DFSPatients who underwent surgical resection (denominator not restated in the abstract)Disease-free survival measured from the time of surgeryfinalmedian follow-up 27.5 monthsBang 2025median 49.3 months; 91% at 12 mo (79.9–100.0%)—Median DFS 95% confidence interval 25.3 months to not reached (upper bound not estimable). Median follow-up for the whole group was 27.5 months (95% CI 22.0-39.2).
Grade 3 adverse eventsAll 40 patients (9 with grade 3 adverse events as printed)finalBang 202522.5%—The abstract does not say whether counts are patients or events, or whether all adverse events or only treatment-related ones are counted.
Grade 4 adverse eventsAll 40 patients (5 with grade 4 adverse events as printed)finalBang 202512.5%—Same caveat as for grade 3 events. The abstract reports that two patients with thymoma died from myocarditis.
Illustrative — modelled from reported values, not trial data. DFS, final.
0%25%50%75%100%020406080100120Months (modelled)
  • Docetaxel, cisplatin and pembrolizumab, surgery, then pembrolizumab (experimental)

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

Data table (modelled values)
MonthsDocetaxel, cisplatin and pembrolizumab, surgery, then pembrolizumab
0100.0%
3164.7%
6241.8%
9227.4%
12317.7%
Primary publication: Bang YH, Noh JM, Lee B, Choi YL, Yang K, Pyo H, et al. Perioperative Pembrolizumab for Locally Advanced Thymic Epithelial Tumors: A Single-Arm, Phase 2 Trial. J Thorac Oncol. 2025;20(12):1829-1842. PubMedDOI

Appears in

Sources

  1. Bang YH, Noh JM, Lee B, Choi YL, Yang K, Pyo H, et al. Perioperative Pembrolizumab for Locally Advanced Thymic Epithelial Tumors: A Single-Arm, Phase 2 Trial. J Thorac Oncol. 2025;20(12):1829-1842.

    PubMedDOI