Targeting PD-1 Classical Hodgkin Lymphoma

Recent evidence suggests that classical Hodgkin lymphoma (cHL) is ideally suited to therapy with checkpoint inhibitors. In Reed-Sternberg cells, the near uniform upregulation of programmed death-ligands 1 and 2 (PD-L1 and PD-L2) result from copy number alterations in chromosome 9p24.1. The clinical responses and correlative studies in the phase 1 studies of nivolumab and pembrolizumab in relapsed and refractory Hodgkin lymphoma have clearly borne out this hypothesis.

In the nivolumab trial reported by Dr. Anas Younes and colleagues, 80 patients with relapsed cHL whose disease recurred following autologous stem cell transplantation and had progressed or failed to respond to brentuximab vedotin were treated with 3 mg/kg of study drug every two weeks until progression. Patients were heavily pretreated, having received a median of four prior therapies. With a median follow-up of approximately nine months, the overall and complete response rates were 66 percent and 9 percent, respectively. Thirty-one patients were enrolled on the pembrolizumab study reported by Dr. Philippe Armand and colleagues. Eligible patients were required to have received prior brentuximab vedotin and undergone autologous stem cell transplantation (ASCT) unless ineligible or declined. The drug was administered at a dose of 10 mg/kg every two weeks until disease progression. More than half of the patients had received five or more prior therapies. 71 percent of patients had failed autologous stem cell transplantation (ASCT). In terms of the activity of pembroluzimab, it was remarkably consistent with that of nivolumab. The overall response rate was 65 percent. Seventy-three percent of patients who had undergone prior ASCT and 44 percent of those who were deemed ineligible for transplantation responded. Complete remissions were seen in 16 percent, 14 percent and 22 percent of patients, respectively. Overall, therapy in both trials was well tolerated with fatigue, infusion reaction and fever being the most common adverse events in the nivolumab study. Hypothyroidism, diarrhea, nausea, and pneumonitis were the most common toxicities with pembrolizumab.

Correlative studies in the nivolumab trial were performed on tissue biopsies from 45 patients. Fluorescence in situ hybridization analyses of Reed-Sternberg cells revealed polysomy, copy gain, and amplification of 9p24.1 in 16 percent, 58 percent, and 27 percent of patients, respectively. Additionally, the JAK-STAT pathway was activated in all samples, as evidenced by staining of phosphorylated STAT3. Interestingly, the investigators demonstrated a correlation between response and the level of 9p24.1 alteration and of PD-L1 expression. None of the patients with the lowest level (polysomy) achieved a complete remission. Of the patients who experienced progressive disease, none had amplification. Additionally, patients who achieved a complete remission had the highest levels of PD-L1 staining and none had polysomy.

In the pembroluzimab study, tissue specimens were evaluable from 16 patients at the time of study enrollment. By immunohistochemical staining, 94 percent expressed PD-L1, and 90 percent of the 10 patients assessed for PD-L2 were positive. Nine patient samples were assessed for T-cell subsets by flow cytometry. Comparing baseline and post–cycle 7 samples, the absolute numbers of T-cells, CD4, CD8, and NK cells all rose during therapy. Additionally, using NanoString technology, paired samples from 19 patients were analyzed. IFN γ-induced genes increased with treatment, as did gene signatures related to T-cell receptor signaling and other immune-related genes. Given the small number of samples, the changes were not predictive of response.