ASH 2018 | Pembrolizumab for R/R PMBCL patients: Results from the phase II Keynote-170 trial and updated data from the phase Ib Keynote-013

On Saturday 1 December 2018, Oral Session 626 took place at the 60^th American Society of Hematology (ASH) Annual Meeting, San Diego, CA. During that session, through Abstract #228 the updated results of the Keynote-013 (KN013) phase Ib trial and the first analysis of its subsequent phase II trial Keynote-170 (KN170), were presented by Philippe Armand from the Dana-Farber Cancer Institute, Boston, MA, USA.

In the primary analysis of the phase Ib KN013 study (NCT01953692), relapsed or refractory (R/R) patients with primary mediastinal large B-cell lymphoma (PMBCL), achieved frequent and durable responses with pembrolizumab. They achieved an objective response rate (ORR) of 41% and a complete response rate of 12%. Here the investigators presented the updated results from this phase Ib trial together with the primary phase II analysis (NCT02576990) on the efficacy and safety of pembrolizumab in R/R PMBCL. Pembrolizumab is a humanized IgG4 anti-PD1 monoclonal antibody. The primary endpoint of KN170 was ORR by blinded independent central review (BICR). Secondary endpoints included, progression-free survival (PFS), overall survival (OS), and safety. Exploratory endpoints included response by Lugano 2014 criteria.

Study designs & baseline characteristics

• KN013 (phase Ib):
  • Data cut-off: 4 April 2018
  • N = 21 (17 evaluable) R/R PMBCL patients, who had failed, were ineligible for, or refused autologous stem cell transplant (ASCT).
  • The initial 10 patients received pembrolizumab 10 mg/kg every two weeks [Q2W]; while the next 11 patients received pembrolizumab 200 mg every three weeks [Q3W] for up to 2 years or until unacceptable toxicity, disease progression or patient/physician withdrawal decision
  • Archival or fresh tumor tissue obtained before pembrolizumab initiation was used for correlative studies
  • Tumor response was assessed with positron emission tomography/computerized tomography (PET/CT) scans according to the International Working Group (IWG) 2007 criteria either:
    • At Week 12 and then every eight weeks for the 10 mg/kg cohort
    • At Week 6 and 12 and then every three weeks for the 200 mg cohort
  • Median patient age (range): 31 (22−62)
  • Median number of prior lines (range): 3 (2−9)
  • Prior radiation: 71% of patients
• KN170 (phase II):
  • Data cut-off: 13 April 2018
  • N = 53 R/R PMBCL, who had relapsed after or were ineligible for ASCT with ≥ 2 lines of prior therapy
  • All patients received pembrolizumab 200 mg Q3W for up to 2 years or until unacceptable toxicity, disease progression or patient/physician withdrawal decision
  • Archival or fresh tumor tissue obtained before pembrolizumab initiation was used for correlative studies
  • Tumor response was assessed by PET/CT scans at Week 12 and then every twelve weeks according to the IWG 2007 criteria
  • Median patient age (range): 33 (20−61)
  • Median number of prior lines (range): 3 (2−8)
  • Prior radiation: 32% of patients

Results

KN170 (phase II) primary analysis:

• Median follow-up (range): 12.5 (0.1−6) months
• Median time to response: 2.8 months
• Response rates by Lugano criteria:
  • Overall response: 43% (n = 23)
  • Complete response (CR): 21% (n = 11)
  • Partial response (PR): 22% (n = 12)
  • Stable disease (SD): 9% (n = 5)
  • Progressive disease (PD): 25% (n = 13)
  • Non-evaluable: 23% (n = 12)
• Response rates by Cheson criteria:
  • Overall response: 45% (n = 24)
  • Complete response (CR): 13% (n = 7)
  • Partial response (PR): 32% (n = 17)
  • Stable disease (SD): 9% (n = 5)
  • Progressive disease (PD): 23% (n = 12)
  • Non-evaluable: 23% (n = 12)
• No patient with CR had relapsed at data cut-off
• 12-month PFS: 38%
• Median PFS: 5.5 (95% CI, 2.8−1) months
• 12-month OS: 58%
• Median OS: not reached (95% CI, 7.3−not reached)

KN013 (phase Ib) updated results:

• Median duration of follow-up (range): 29.1 (0.6−6) months
• Median time to response: 2.7 months
• Two patients in CR at two years, remained in CR after a further 12 and 18 months of follow-up therapy
• 12-month PFS: 47%
• Median PFS: 10.4 (95% CI, 3.4−not reached) months
• 12-month OS: 65%
• Median OS: 31.4 (95% CI, 4.9−not reached) months

Safety

KN170 (phase II) primary analysis:

• Treatment-emergent adverse events (TEAEs) of all grades occurred in 57% of patients (n = 30)
• Grade 3−4 TEAEs occurred in 23% of patients (n = 12)
• Most common Grade 3−4 TEAEs were:
  • Neutropenia: 13%
  • Febrile neutropenia: 2%
  • Increased alanine aminotransferase (ALT): 2%
  • Increase aspartate aminotransferase (AST): 2%
  • Clostridium difficile infection: 2%
  • Pneumonia: 2%
  • Tumor flare: 2%
  • Venous thrombosis: 2%
• Immune-mediated AEs of any grade occurred in 11% (n = 6) of patients
• Grade 3−4 immune-mediated AEs occurred in one patient (pneumonitis)

KN013 (phase Ib) updated results:

• Treatment-emergent adverse events (TEAEs) of all grades occurred in 71% of patients (n = 15)
• Grade 3−4 TEAEs occurred in 24% of patients (n = 5)
• Most common Grade 3−4 TEAEs were:
  • Neutropenia: 14%
  • Febrile neutropenia: 5%
  • Increased ALT: 5%
  • Fatigue: 5%
  • Hyponatremia: 5%
• Immune-mediated AEs of any grade occurred in 19% (n = 4) of patients
• Grade 3−4 immune-mediated AEs occurred in one patient (myositis)

 Conclusions

• Pembrolizumab leads to frequent and durable response in heavily pre-treated R/R PMBCL patients
• The median DoR was not reached with a median follow-up of 29.1 months in KN013 or with a median follow-up of 12.5 months in KN170
• At least 50% of patients were alive at one year after treatment in both studies
• Pembrolizumab has a manageable safety profile in R/R PMBCL patients
• The presented data provided the basis for the accelerated approval of pembrolizumab for R/R PMBCL patients from the Food and Drug Administration (FDA), US on 13 June 2018