Immunotherapy of hematological cancers: PD-1 blockade for the treatment of Hodgkin's lymphoma

Abstract

The blockade of immunological checkpoints has been successfully employed for the treatment of various solid neoplasms including melanoma, mesothelioma, non-small cell lung carcinoma, and renal cell carcinoma. A recent study indicates that the vast majority of patients with advanced, heavily pretreated Hodgkin's lymphoma (HL) also respond to a monoclonal antibody targeting programmed cell death 1 (PDCD1, best known as PD-1). Thus, checkpoint blockers may soon become part of our therapeutic armamentarium against hematological tumors. This would be particularly important as it would spare (at least some) patients the deleterious toxic effects of combinatorial chemotherapies and bone marrow transplantation. We anticipate that the realm of immunotherapy will eventually conquer vast portions of the territory that now belongs to hematological malignancies.

Keywords:

• CTLA4
• immunostimulatory monoclonal antibodies
• ipilimumab
• PD-L1
• pembrolizumab
• pidilizumab

The end of 2014 has been marked by the two major achievements in the field of immunotherapy for hematological malignancies.

First, the US Food and Drug Administration (FDA) approved blinatumomab, a CD19- and CD3-targeting bispecific T-cell engager (BiTE) developed by Amgen (Thousand Oaks, CA, US), for the treatment of acute B-cell lymphoblastic leukemia.¹ Blinatumomab is the first BiTE ever approved for use in humans, an important victory over the adoptive transfer of chimeric antigen receptor (CAR)-expressing T cells, which is logistically and economically way more demanding.^2,3

Second, Ansell and colleagues (Mayo Clinic, Rochester, MN, US) published in the New England Journal of Medicine preliminary results from a hitherto ongoing Phase I clinical study testing the safety and efficacy of various immunomodulatory monoclonal antibodies (mAbs) in patients with lymphoma or multiple myeloma (NCT01592370).^4,5 The observations by Ansell and collaborators demonstrate for the first time that nivolumab, a checkpoint-blocking mAb targeting PD-1 that is currently approved for the treatment of advanced melanoma,⁶ is highly effective in a majority of patients with relapsed or refractory HL.⁷

In HL patients, lymph nodes enlarge as they are massively colonized by malignant cells known as Hodgkin and Reed–Sternberg cells as well as by a reactive cell infiltrate composed of T lymphocytes and other leukocytes in variable proportions.^8,9 Patients affected by HL receive a first-line chemotherapeutic cocktail that has never been considered as an immunotherapeutic regimen.¹⁰ However, the drugs that have been used to treat HL so far exert anticancer effects that rely (at least in part) on the (re)activation of anticancer immunosurveillance.^11,12 Indeed, doxorubicin (a DNA-damaging agent) and bleomycin (a glycopeptide antibiotic) are bona fide inducers of immunogenic cell death (ICD);^13,14 vincristine (a microtubular poison) stimulates antigen presentation by dendritic cells,¹¹ and cyclophosphamide (an alkylating agent) not only promotes ICD, but also favors the establishment of an immunostimulatory tumor microenvironment by influencing the myeloid cell infiltrate.¹⁵ Along similar lines, external beam radiation therapy can induce ICD, at least in some circumstances,^16,17 and autologous hematopoietic stem-cell transplantation (HSCT) may reset the immune system and promote the establishment of a tumor-reactive state.¹⁸ Finally, since August 2011, patients with HL progressing after HSCT can be treated with brentuximab vedotin, an antibody-drug conjugate directed against CD30, which is expressed by Hodgkin and Reed-Sternberg cells.¹⁹ Of note, approximately 20% of HL patients fail to respond to all of these therapeutic options. Moreover, systemic chemotherapy, radiotherapy, as well as HSCT are associated with significant toxicity in a fraction of patients. This leaves space for improvement of the protocols that are routinely employed in the clinical management of HL.

The paper by Ansell et al. reports data from a relatively small cohort of subjects with HL (n = 23), most of whom relapsed following one (or more) of the regimens described above, including brentuximab vedotin-based immunotherapy. These patients received nivolumab at a dose of 3 mg/kg at week 1, week 4, and then every 2 weeks until disease progression or complete response or for a maximum of 2 years. Strikingly, 4 individuals (17%) experienced a complete response, 16 (70%) exhibited a partial response and 3 (13%) had stable disease.⁷ An overall response rate (ORR) as high as that documented by this study (>85%) has never been reported before among patients with hematological cancers treated with checkpoint blockers, creating a precedent in the field. It can be anticipated that, if confirmed in independent studies involving larger patient cohorts, these results will accelerate the approval of nivolumab for the treatment of relapsed or refractory HL. Subsequent clinical trials will have to determine whether nivolumab might also be employed as a first-line therapeutic option. This would be particularly relevant as it would spare patients from the short- and long-term side effects of chemotherapy, radiotherapy and autologous HSCT.

Importantly, the pronounced effects of nivolumab on HL patients may be directly linked to the molecular features of this disease. Indeed, HL is often associated with the amplification of a chromosomal region (9p24.1) that encompasses the genes coding for the two main ligands of PD-1, i.e., CD274 (best known as PD-L1) and PDCD1 ligand 2 (PDCD1LG2, best known as PD-L2) as well as the gene coding for Janus kinase 2 (JAK2), a kinase involved in the transactivation of CD274 and PDCD1LG2.²⁰ Such a combination of molecular defects results in the robust overexpression of both PD-L1 and PD-L2,²⁰ to which T lymphocytes that infiltrate the lymph nodes of HL patients are particularly sensitive (as these T cells express high levels of PD-1).²¹ In this context, blocking the interactions of PD-1 with its ligands may alleviate the functional inhibition of T lymphocytes, hence (re)activating a tumor-targeting immune response, and/or promote the demise of cancer cells (which is normally prevented by PD-L1 retrosignaling). The precise mechanisms through which nivolumab induces clinical responses in HL patients, however, remain to be determined.

Irrespective of these unknowns, the study by Ansell and co-authors may open a new therapeutic perspective for patients with HL. Previous studies suggested that checkpoint blockers may be beneficial to individuals affected by various forms of non-HL. In 2013, Armand et al. reported an ORR of 51% among diffuse large B-cell lymphoma patients treated with pidilizumab, another (hitherto experimental) PD-1-targeting mAb, after autologous HSCT.²² More recently, Westin and colleagues documented an ORR of 66% among subjects with relapsed follicular lymphoma treated with pidilizumab in combination with the anti-CD20 mAb rituximab.²³ It will be interesting to see whether pembrolizumab, another (FDA-approved) checkpoint-blocking mAb specific for PD-1, and/or mAbs that target PD-L1, are equally effective in patients with HL or other forms of lymphoma. Several clinical trials testing this possibility are under way (source: https://clinicaltrials.gov).

One obvious particularity of lymphomas consists in the fact that neoplastic cells are preponderantly found in lymphoid organs, i.e., in a microenvironment that is specifically organized for the elicitation of immune responses. It remains to be seen whether leukemias, which develop in a relative more diffuse manner, respond to checkpoint blockers as lymphomas do. To clarify this issue, several clinical trials evaluating the safety and efficacy of nivolumab (NCT02011945, NCT02275533), pidilizumab (NCT01096602) and pembrolizumab (NCT02332980) in leukemia patients have recently been initiated (source: https://clinicaltrials.gov).