Questions persist regarding mechanisms of PD-L1 overexpression in lymphoid malignancies, patient selection, biomarkers to guide therapy and predict response, and identification of agents to use in combination with PD-1/PD-L1 inhibitors.
Limited clinical evaluation suggests a potentially major role for immune checkpoint inhibition in the treatment of lymphomas, but early success has also raised many questions, according to a presentation at the 2015 Society of Hematologic Oncology (SOHO) annual meeting.
A small clinical trial of the PD-1 inhibitor Opdivo (nivolumab) in relapsed/refractory Hodgkin lymphoma showed a high response rate and durable remissions in a difficult patient population, said Margaret Shipp, chief of Hematologic Neoplasia at Dana-Farber Cancer Institute. However, questions persist regarding mechanisms of PD-L1 overexpression in lymphoid malignancies, patient selection, biomarkers to guide therapy and predict response, and identification of agents to use in combination with PD-1/PD-L1 inhibitors.
The human immune system comprises multiple positive and negative modulators of immune response. Under normal circumstances, negative modulators serve several purposes: downregulation of the immune response after elimination of a disease threat, limiting the strength of response to prevent damage to normal tissue, and maintaining immune tolerance, Shipp noted.
Over the past several years, emerging data have indicated that lymphomas and multiple other types of malignancies co-op normal coinhibitory signaling pathways to evade effective antitumor immune response.
Investigation of immunotherapeutic approaches to lymphoid malignancies received a serendipitous boost from another line of research. During studies of the molecular signatures of different types of lymphoma, Shipp and colleagues discovered a number of similarities between primary mediastinal large B-cell lymphoma (MLBCL) and classical Hodgkin lymphoma. The similarities included transcriptional profiles, constitutive NF kappa-B activation, chromosome 9 copy gain, prominent inflammatory/fibrotic immune infiltrate, and common clinical features, such as young age and localized disease at diagnosis.
“Hodgkin lymphoma is somewhat of a paradox in that less than 5% of the cells in the primary tumor are actually malignant cells,” says Shipp. “Around those few malignant cells is an extensive inflammatory immune-cell infiltrate.”
“Another paradox is that despite this infiltrate, there is no compelling evidence of an effective antitumor immune response,” she continued. “We were interested in trying to determine whether there were genetic bases for the lack of an effective immune response in this setting.”
Investigators performed an extensive integrative analysis to identify recurrent copy number alterations and candidate driver genes associated with classical Hodgkin lymphoma. They focused on transcripts that were most closely associated with recurrent copy gain.
Studies of MLBCL and Hodgkin cell lines identified recurrent copy gain at chromosome 9p24.1 as one of the most reproducible copy alterations. The transcripts most closely associated with recurrent copy gain were PD-1 ligands, PD-L1, and PD-L2. Similar findings emerged from studies of tumor specimens from patients with MLBCL and Hodgkin lymphoma.
The PD-L1 and PD-L2 copy gain at 9p24.1 and subsequent binding of PD-1 receptors on T-cells is associated with multiple effects that contribute to the malignant process. These include dephosphorylation of proximal signaling kinases, reduced signal, reduced T-cell activation, and reversible T-cell exhaustion.
PD-L1 and PD-L2 copy gain also is associated with copy number-dependent increase in Janus kinase (JAK)-2, leading to increased expression and activation of JAK-2 and activation and phosphorylation of downstream STAT. JAK/STAT signaling induces further increases in PD-L1 and PD-L2 transcription.
With the wealth of information from preclinical studies, Shipp and colleagues proceeded to clinical evaluation. They conducted a study involving 23 patients with relapsed/refractory Hodgkin lymphoma. Three-fourths of the patients had prior autologous stem cell transplantation plus Adcetris (brentuximab vedotin). One-third of the patients had received six or more prior therapies and a similar number had received four or five prior regimens.
Treatment with the PD-1 inhibitor Opdivo led to an overall response rate of 87 percent, including complete responses in 26 percent. After a median follow-up duration of 86 weeks, the median response duration had yet to be reached. Responses were often rapid and in some cases dramatic.
“It’s important to keep in mind that the therapy is not targeting the tumor but instead is targeting the efficacy of the antitumor immune response,” says Shipp.
Investigators have since developed a fluorescence in-situ hybridization assay for PD-L1 and PD-L2 alterations. Tests have demonstrated the assay’s ability to associate copy gain and amplification with increased expression of the ligands on tumor cells.
The FDA granted breakthrough therapy status to Opdivo as a treatment for patients with Hodgkin lymphoma following autologous stem cell transplant and Adcetris. Enrollment at 35 centers participating in a registration trial in this setting is almost complete, says Shipp. A second registration trial with a different PD-L1 inhibitor has also begun.
Studies have also begun to identify other lymphoid malignancies that might be amenable to immune checkpoint inhibition. The work has led to identification of multiple mechanisms by which PD-1 ligand overexpression may occur. In addition to gene amplification, possibilities include chromosomal translocation, viral infection, and as-yet undetermined mechanisms associated with subtypes of large B-cell lymphoma and high-risk diffuse large B-cell lymphoma.
The work to date has raised a number of questions about how to make the best use of immune checkpoint inhibitors in the treatment of lymphoid malignancies, said Shipp. Considerations include identification of optimal agents for combination therapy to improve response rates; determination of the optimal setting for use of combinations; methods to predict response to PD-1 blockade alone or in combination; whether responses are maintained off therapy or can be restored with retreatment; and identification of other lymphoid malignancies dependent on PD-1 signaling and other immunomodulatory pathways.