Zinzani PL, et al. ASH 2017:2833

Efficacy and safety of pembrolizumab in patients with R/R PMBCL: An updated interim analysis of the phase II KEYNOTE-170 study


Prognosis is poor for patients with relapsed or refractory (R/R) primary mediastinal B-cell lymphoma (PMBCL). Treatment options are especially limited for patients who have failed or who are ineligible for autologous stem cell transplantation (ASCT).1–3 In a phase Ib trial, pembrolizumab demonstrated promising antitumour activity with a manageable safety profile in patients with R/R PMBCL.4 Updated results of the phase II KEYNOTE-170 trial, which is investigating the efficacy and safety of pembrolizumab in this patient population, were presented at the 2017 ASH Annual Meeting.5

Study design

  • KEYNOTE-170 is an ongoing, multicentre, phase II study investigating the efficacy and safety of pembrolizumab in patients with R/R PMBCL and in patients with R/R Richter syndrome.
    • The present analysis included only the patients with PMBCL.
  • The primary endpoint was overall response rate (ORR) by blinded, independent central review per International Working Group 2007 criteria.
  • Secondary endpoints included duration of response (DOR), safety, and tolerability.
  • Response assessments (by positron emission tomography/computed tomography scan) occurred at Week 12 and every 12 weeks thereafter.
  • The efficacy population was comprised of the first 29 treated patients, all of whom had reached the Week 24 assessment or discontinued earlier by the time of database lock.
  • All patients who received at least one dose of pembrolizumab were included in the safety population.
  • A sample size of 50 patients was planned to allow 88% power to detect an ORR of 35% or higher with pembrolizumab compared with a 15% fixed control rate, using the exact binomial test.

Study design

Key findings

Baseline characteristics and disposition

  • A total of 53 patients were enrolled in the study.
  • Median age was 33 years (range: 20–61).
  • The majority of patients were female (57%), transplant ineligible (74%), had an Eastern Cooperative Oncology Group performance status of 1 (57%), and were positive for programmed death-ligand 1 (PD-L1) (64%).
  • Patients received a median of three prior lines of therapy (range: 2–8) and had received rituximab as a prior treatment.
  • Of all treated patients (n = 53), 19 (36%) are still undergoing treatment and 34 (64%) have discontinued treatment.
  • Reasons for discontinuation were:
    • Progressive disease (n = 14);
    • Clinical progression (n = 12);
    • Adverse event (n = 4);
    • Physician decision (n = 3); and
    • Complete response (n = 1).
  • In the efficacy population (n = 29), 22 patients (76%) were positive for PD-L1, one patient (3%) was PD-L1 negative, and six patients (21%) had missing data.


  • At the database cutoff of August 15, 2017, median duration of follow-up was 10.5 months (range: 0.1–17.7) in the efficacy population (n = 29).
  • ORR was 41%, including 24% with complete responses and 17% with partial responses. (Table 1)
  • Median time to response was 2.8 months (range: 2.4–5.5).
  • Best responses in key patient subpopulations are summarized in Table 2.
  • Median duration of response was not reached (range: 1.1+ to 13.6+ months).
    • At the time of analysis, seven responses were ongoing and no responders had progressive disease.
    • In three patients, partial response was converted to a complete response over time.
    • Two patients underwent ASCT while in remission.
  • A reduction in target lesions was observed in 16 of 22 evaluable patients (73%).

Table 1. Antitumour activity in patients with R/R PMBCL treated with pembrolizumab

Table 2. Best response in key patient subpopulations


  • Median duration of pembrolizumab treatment in all patients was 3.5 months (range: 0.03–17.2).
  • Treatment-related adverse events (AEs) are summarized in Table 3.
  • Immune-related AEs are summarized in Table 4.

Table 3. Treatment-related adverse events

Table 4. Immune-mediated adverse events

Key conclusions

  • In patients with R/R PMBCL, including heavily pretreated patients, pembrolizumab had:
    • A high response rate;
    • An ability to induce durable responses, with some responses deepening over time; and
    • A manageable safety profile consistent with prior experience of pembrolizumab in other tumour types.

References: 1. Martelli M, Di Rocco A, Russo E, et al. Primary mediastinal lymphoma: diagnosis and treatment options. Expert Rev Hematol 2015;8(2):173–86. 2. Zinzani PL, Pellegrini C, Chiappella A, et al. Brentuximab vedotin in relapsed primary mediastinal large B-cell lymphoma: results from a phase 2 clinical trial. Blood 2017;129(16):2328–30. 3. Jacobsen ED, Sharman JP, Oki Y, et al. Brentuximab vedotin demonstrates objective responses in a phase 2 study of relapsed/refractory DLBCL with variable CD30 expression. Blood 2015;125(9):1394–402. 4. Zinzani PL, Ribrag V, Moskowitz CH, et al. Safety and tolerability of pembrolizumab in patients with relapsed/refractory primary mediastinal large B-cell lymphoma. Blood 2017;130(3):267–70. 5. Zinzani PL, Thieblemont C, Melnichenko V, et al. Efficacy and safety of pembrolizumab in patients with relapsed or refractory primary mediastinal large B-cell lymphoma: an updated interim analysis of the phase 2 KEYNOTE-170 study. ASH Annual Meeting Abstracts 2017:2833.