Checkpoint inhibition in CNS lymphoma

Beth Christian

In this issue of Blood, Nayak et al1 report the clinical outcomes for 5 patients with relapsed or refractory primary central nervous system (CNS) lymphoma (PCNSL) and primary testicular lymphoma (PTL) with secondary central nervous system involvement who were treated with nivolumab, a programmed cell death 1 (PD-1) inhibitor.

One of the most challenging clinical scenarios to manage in patients with diffuse large B-cell lymphoma is primary or secondary CNS involvement because of the debilitating nature of the diagnosis and the toxicities of the currently available therapies. Traditional treatment of CNS lymphoma includes high-dose methotrexate–based induction followed by a variety of consolidation strategies, including radiation and autologous stem cell transplant. High-dose methotrexate is difficult to administer, requiring both intensive monitoring and access to the necessary supportive care that is not readily available in all settings. The treatment approach for patients with advanced age, poor performance status, or impaired organ function is generally palliative. In the case of PTL, concurrent CNS and systemic involvement presents even greater clinical challenges. Treatment is also associated with the potential for significant acute chemotherapy-associated toxicity and long-term neurotoxicity. Even with intensive therapy, the prognosis of primary and secondary CNS lymphoma remains poor.2,3

Recently, novel targeted approaches have shown promise in primary and secondary CNS lymphoma. Phase 1 and 2 studies of ibrutinib, a Bruton tyrosine kinase inhibitor, have demonstrated overall response rates ranging from 56% to 75%,4,5 and ibrutinib has also been successfully incorporated into a combination chemotherapy regimen.6 Lenalidomide, an immunomodulatory agent, has also demonstrated preliminary efficacy in combination with rituximab in CNS lymphoma.7,8

The shared genetic characteristics of PCNSL and PTL have been previously described by the authors9 and support exploration of other novel targeted therapies in these diseases. PCNSL and PTL exhibit unique genetic alterations in comparison with systemic diffuse large B-cell lymphoma, including a high frequency of 9p24 copy gain and increased PD-1 ligand expression, as well as chromosomal translocations resulting in PD-L1/PD-L2 deregulation.9 Based on these genetic characteristics, the authors hypothesized that PD-1 blockade was a rational therapeutic approach in PCNSL and PTL.

In the current case series, the authors report their clinical experience administering nivolumab off-label to patients with relapsed or refractory CNS lymphoma who had exhausted standard therapy. Reported toxicities in this series were consistent with the expected side effect profile of nivolumab. Notably, none of the patients experienced “pseudoprogression,” an immune-mediated phenomenon resulting in a transient increase in tumor burden. All 5 patients in this series had evidence of durable responses lasting >1 year, with 3 patients remaining without evidence of progression, exceeding expected survival in this setting.

PCNSL and PTL frequently affect older patients and oftentimes result in compromised performance status, as illustrated in this series. In a disease where standard therapy can be associated with prohibitive short- and long-term risk, this case series has significant clinical implications. These preliminary data, along with the genetic features that result in immune evasion through 9p24.1/PD-L1/PD-L2 copy-number alterations and translocations,9 support evaluation of PD-1 blockade in PCNSL and PTL in the setting of a clinical trial. A phase 2 study of nivolumab is currently underway (NCT02857426). In the future, combinations of novel targeted therapies both in the relapsed setting and incorporated into front-line regimens may potentially alter the historically poor prognosis of PCNSL and secondary CNS involvement of PTL.


  • Conflict-of-interest disclore: B.C. received research funding from Bristol-Myers Squibb, Merck, Genentech, Celgene, Seattle Genetics, Immunomedics, Acerta, Janssen, and Pharmacyclics and is on the Arzerra data safety monitoring board.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.