in lymphoma Catherine Hildyard Haematology Senior Registrar Oxford - PowerPoint PPT Presentation

Immune checkpoint inhibitors in lymphoma Catherine Hildyard Haematology Senior Registrar Oxford University Hospitals NHS Foundation Trust

Aims • How immune checkpoint inhibitors work • Success of immune checkpoint inhibitors in Hodgkin lymphoma • Toxicity profiles of immune checkpoint inhibitors • Immune checkpoint inhibitors in non-Hodgkin lymphoma?

How immune checkpoint inhibitors work Priming phase : CTLA-4 and PD-1 Effector phase : PD-1 Hude et al. (2017) Haematologica

Pre-determined genetic susceptibility of cHL to PD-1 blockade? PD-L1 PD-L2 Roemer et al. (2016) J Clin Oncol

Unmet clinical need • Median survival of r/r cHL after failure of ASCT = 2 years 1 • Brentuximab has a 75% response rate after failure of ASCT but with progression free survival of only 5.6 months 2 • Response rate of brentuximab in transplant ineligible patients is much lower at 30% 3 1. Von Tresckow et al. (2014) Leuk Lymphoma 2. Younes et al. (2012) J Clin Oncol 3. Forero-Torres et al. (2012) Oncologist

Nivolumab in Hodgkin: Checkmate 039 • Phase 1 study (Checkmate-039) • 23 patients • Relapsed / refractory cHL • Most had had ASCT and brentuximab • 87% had 3 or more lines of Rx Ansell et al. (2014) NEJM

Nivolumab in Hodgkin: Checkmate 039 • Every patient had some tumour reduction • 87%: partial response + complete response • Only 17% (4 pts) had complete response Ansell et al. (2014) NEJM

Nivolumab in Hodgkin: Checkmate 205 Relapsed cHL after autologous (auto)-HSCT nivolumab monotherapy Cohort A Cohort B Cohort C n = 63 n = 80 n = 100 Primary endpoint • ORR Nivolumab 3 mg/kg Brentuximab Brentuximab Brentuximab IV Q2W vedotin naïve vedotin vedotin before Treatment until Additional after and/or after disease progression or endpoints auto-HSCT auto-HSCT unacceptable toxicity • DOR • Duration of CR/PR Patients could elect to discontinue nivolumab and • PFS by IRC proceed to allogeneic SCT • OS • Safety 19 months 23 months 16 months f/up f/up f/up Armand et al. (2018) J Clin Oncol

Best Overall Response by cohort BV before and/or BV after auto- BV naïve after Overall HSCT (Cohort A) auto-HSCT (Cohort B) n=63 (Cohort C) n=243 n=80 n=100 Objective response, % 65 68 73 69 Best overall response, % Complete remission 29 13 12 16 Partial remission 37 55 61 53 Stable disease 24 21 15 19 Progressive disease 11 8 10 9 Unable to determine 0 4 2 2 • >95% of patients had reduction in tumour burden • Responses were similar irrespective of BV treatment sequence

Duration of Response by Best Overall Response 1.0 0.9 Proportion of patients in response 0.8 0.7 CR: Median: 20 months 0.6 0.5 0.4 0.3 PR: Median: 13 months 0.2 0.1 0.0 0 3 6 9 12 15 18 DOR (months)

Outcomes After Allo-SCT (n=44) Progression-free survival 1.0 0.9 100-day rate 0.8 87% 6-month rate Probability of PFS from 0.7 subsequent allo-HSCT 82% 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0 50 100 150 200 PFS from allo-SCT (days)

Toxicity post allo-SCT • But, 5% (2 pts) hyper-acute GVHD

Similar trials with pembrolizumab: Keynote-087 study Cohort 1 Cohort 2 Cohort 3 n = 69 n = 81 n = 60 SCT-naïve Failed BV treated BV naïve salvage post-ASCT post- chemo & ASCT BV Chen et al. (2017) J Clin Oncol

Similar trials with pembrolizumab: Keynote-087 study Cohort 1 Cohort 2 Cohort 3 n = 69 n = 81 n = 60 SCT-naïve Failed BV treated BV naïve salvage post-ASCT post- chemo & ASCT BV Chen et al. (2017) J Clin Oncol

Similar trials with pembrolizumab: Keynote-087 study Cohort 1 Cohort 2 Cohort 3 n = 69 n = 81 n = 60 SCT-naïve Failed BV treated BV naïve salvage post-ASCT post- chemo & ASCT BV Chen et al. (2017) J Clin Oncol

Patients with a drug- related AEs (≥10%), serious Overall population AEs (≥1%), or AEs leading to discontinuation (N = 243) (≥1%) Grade 3 – 4 Any grade Drug-related AE, % Fatigue 23 1 Diarrhea 15 1 Infusion-related reaction 14 0 Rash 12 1 Nausea 10 0 Pruritus 10 0 Drug-related serious AE, % Infusion-related reactions 2 <1 Pneumonitis 1 0 Drug-related AEs leading to discontinuation, % Pneumonitis 2 0 Autoimmune hepatitis 1 1

Case study • 35 year old man • cHL diagnosed 2y ago, stage IIIA • 6xABVD – PR • 3 months later – progressive disease • 2xIVE – very good PR • 1xIVE and ASCT • 8 months later – progressed • BV 3 courses: PR • BV further 2 courses: PD prior to allo-SCT • Commenced pembrolizumab on trial

PD-1 inhibitors are not without significant toxicity Shortly after 3 rd infusion: • Admitted with urinary retention • Ascending paraesthesia • Developed paraplegia and tingling in both arms • Intractable hiccoughs • MRI performed Rx commenced: high dose steroids, IVIG and plasma exchange

Phase 1 data for PD-1 inhibitors in NHL • Phase 1 trial of nivolumab • r/r B cell and T cell NHL Lesokhin et al. (2016) J Clin Oncol

Preliminary phase 2 data for PD-1 inhibitors in DLBCL: checkmate 0139 • Nivolumab in DLBCL relapsed post auto-SCT or ineligible for auto-SCT • 161 patients ASCT failed/ ineligible ORR (%) ASCT failed (n= 87) 10.3 ASCT ineligible (n=34) 2.9 ClinicalTrials.gov Identifier: NCT02038933

Why are PD-1 inhibitors less effective in DLBCL than cHL? Green et al. (2010) PMBCL Blood

PMBCL – a promising subtype? • Pembrolizumab for up to 2 years or until disease progression/unacceptable toxicity • 81% patients had decrease in target lesions • OR 41%, CR 12% • DOR 2.3 to 22.5 months Zinzani et al. (2017) Blood

PCNSL and PTL • Copy number gain and PD-L1/2 over-expression in PCNSL and PTL Nayak et al. (2017) Blood

Epstein-Barr virus (EBV)+ NHL • Increased PD-L1 expression in EBV+ NHL: – PTLD: 60% – Plasmablastic lymphoma: 44% – Primary effusion lymphoma: 50% – Extra-nodal NK/T-cell lymphoma: 67%

NK/T-cell lymphoma and PD-1 inhibition • 7 patients with r/r NKTCL treated with pembrolizumab • All patients responded: 5 CR, 2 PR • 5 patients remained in remission after a median f/u of 6 months • Radiological CR correlated with clearance of EBV DNA • “Pseudo - progression” seen in some patients Kwong et al. (2017) Blood

Enhancing efficacy of PD-1 inhibitors with combination therapy? • BV-treated tumour cells may undergo “immunogenic cell death” 1 • Similar responses can also be induced by chemotherapy and radiotherapy • Checkmate 436 study 2 : nivolumab + brentuximab in r/r DLBCL and PTCL 1. Gardai et al. (2015) Cancer Res (abstract) 2. ClinicalTrials.gov Identifier: NCT02581631

Checkmate 436 case study • 67 year old woman with PTCL: no response to 4x CHOP-14 • Escalated to IVEx2: CR, consolidated by autograft • Relapse 14 months later: severe itch, fatigue, cough, diarrhoea • Symptoms improved after 1 cycle of checkmate 436

PET scans pre and post checkmate 436

Summary • PD-1 inhibitors are very effective in cHL: • Success both in relapse post-ASCT and in transplant ineligible patients • Durable responses with CR and even with PR • Different toxicity profile from conventional chemotherapy • Some questions remain to be answered • PD-1 inhibitors have not yet found a niche in NHL: • Low activity in r/r DLBCL as monotherapy – ?role for combination therapy • Some subtypes may be more responsive (PMBCL, PCNSL/PTL, EBV+)

Thank you Questions? catherine.hildyard@ouh.nhs.uk

PDL1 expression as a biomarker? Younes et al. (2016) Lancet Oncol