A novel class of inhibitors may hold some promise for boosting responses to checkpoint inhibitors and for sensitizing poorly immunogenic tumors, such as pancreatic cancer, to immunotherapy. The drug targets semaphorin 4D (SEMA4D), a glycoprotein expressed on the cell membranes of many tumor types. SEMA4D is associated with an immunosuppressive tumor microenvironment, and antibodies that inhibit it seem capable of “flipping” the microenvironment toward a more immunogenic state.

At the 2020 ASCO-SITC Clinical Immuno-Oncology Symposium, interim results of a phase Ib/II clinical trial were reported in advanced non–small cell lung cancer (NSCLC), as were preclinical studies in pancreatic cancer—a tumor that is immunologically “cold” and has been notoriously difficult to treat with immunotherapy.^1,2 Early signs are encouraging.

CLASSICAL–Lung Combination Trial

Michael Shafique, MD

The novel combination of the anti-semaphorin antibody pepinemab plus the programmed cell death ligand 1 (PD-L1) inhibitor avelumab yielded responses in patients with advanced NSCLC. Durable responses were observed both in patients who were naive to and those resistant to immunotherapy, according to interim results from the phase Ib/II CLASSICAL–Lung Combination trial presented by Michael Shafique, MD, of the Department of Thoracic Oncology at Moffitt Cancer Center, Tampa.¹

SEMA4D binds to plexin receptors located on myeloid cells in the tumor microenvironment, resulting in cell migration and differentiation and creating a generally protumoral microenvironment. Blockade of SEMA4D with pepinemab is believed to shift the balance of chemokines and suppressor cells to enhance infiltration and activity of various antitumoral T cells, dendritic cells, and antigen-presenting cells. Anti-SEMA4D antibodies have been shown to be synergistic with checkpoint inhibitors in preclinical studies.

This phase Ib/II trial evaluated pepinemab plus avelumab in 62 patients with advanced NSCLC who had disease progression on prior chemotherapy and/or anti–programmed cell death protein 1 (PD-1) antibodies. The population included a dose-escalation cohort of 12 patients who were naive to anti–PD-1/PD-L1 agents and a dose-expansion cohort of 50 patients, both immunotherapy-naive and immunotherapy-pretreated, treated with pepinemab at 10 mg/kg. Both cohorts received pepinemab with avelumab at 10 mg/kg given every 2 weeks. Approximately 60% of patients had some degree of expression of PD-L1, though high expressors were rare.

In general, the combination was very well tolerated. The most frequent adverse events were grade 1 or 2 fatigue, chills, and fever. No grade 5 events were attributable to the combination, and the overall immunogenicity of the combination was not concerning, he said.

Of 21 evaluable immunotherapy-naive patients, 5 (23%) achieved partial responses; the disease control rate was 81%. Three patients remained on study for more than 1 year, and two are currently still receiving the combination. Among 29 evaluable immunotherapy-pretreated patients, 2 (7%) responded; the disease control rate was 59%. Three patients are still on active treatment, one of whom has been on therapy for more than 1 year.

Examination of pretreatment and on-treatment biopsies (of the same lesion) in patients with a partial response or stable disease showed a “pretty drastic reduction in viable tumor,” Dr. Shafique said. “Many patients had absent tumor on these repeat treatment biopsies.”

“Clinical response or disease stabilization was observed in the majority of patients, despite low PD-L1 expression.”

— Michael Shafique, MD

Tweet this quote

Of the five patients who remain on treatment, three were immunotherapy-refractory. “They enjoyed partial responses, and we seemed to be able to halt their disease progression to some degree,” he said.

“Clinical response or disease stabilization was observed in the majority of patients, despite low PD-L1 expression,” he added. Some 71% of all subjects with either a partial response or stable disease were reported to have negative or low-positive PD-L1 levels.”

In an exploratory analysis, CD8 density correlated with response and stable disease; after treatment, CD8-positive T cells increased within the tumor. This supports the mechanism of action being a reversal of the myeloid-induced suppression in the microenvironment and an improvement in T-cell infiltration and activity, the researchers said.

Sensitizing Pancreatic Cancer

Luis I. Ruffolo, MD, a surgical resident at the University of Rochester, New York, and fellow in the Center for Tumor Immunology Research there, described preclinical work with SEMA4D inhibition, aimed at sensitizing pancreatic tumors to immunotherapy.²

Luis I. Ruffolo, MD

Dr. Ruffolo noted that, in pancreatic tumors, normal quiescent tissue is largely replaced by desmoplasia dominated by collagen deposition and a dense immune infiltrate. This infiltration of immune cells is largely composed of myeloid-suppressive populations. As a result, checkpoint inhibitors have been largely ineffective, with response rates of 0% in a number of trials, even with combinations of PD-L1 and cytotoxic T-lymphocyte–associated protein 4 (CTLA-4) inhibitors.

Semaphorin 4D is robustly overexpressed in pancreatic cancers, and the SEMA4D ligand co-localized on lymphocytes and tumor-associated macrophages in the microenvironment. Blockade of this axis has shown the ability to “flip” the tumor microenvironment, he said, decreasing the amount of suppression by myeloid subpopulations and allowing an influx of CD8-positive effector T cells.

In a mouse model of pancreatic cancer, Dr. Ruffolo and colleagues gave FOLFIRINOX (leucovorin, fluorouracil, irinotecan, oxaliplatin) with and without combination immunotherapy (anti–PD-1 and anti–CTLA-4) plus an anti-SEMA4D monoclonal antibody. The mice lived longest when treated with the triple combination of FOLFIRINOX, combination checkpoint blockade, and anti-SEMA4D antibody, compared to FOLFIRINOX alone, FOLFIRINOX plus immunotherapy, or FOLFIRINOX plus anti-SEMA4D (P < .02).

In a subsequent experiment in which FOLFIRINOX was not given, mice receiving checkpoint blockade plus anti-SEMA4D had smaller tumors compared to other cohorts and also showed a robust infiltrate of CD8-positive cells (P = .03), “which is not typical of these tumors,” he noted.

The conclusion from these preclinical experiments was that ­SEMA4D blockade augments dual-checkpoint therapy in the context of FOLFIRINOX, according to Dr. Ruffolo.

“We are seeking to translate this novel concept in pancreatic cancer into a first-line treatment for metastatic patients with a FOLFIRINOX backbone and an anti–PD-1/PD-L1 agent in combination with pepinemab, a SEMA4D monoclonal antibody,” he said. Comparison of baseline and on-treatment biopsies will reveal whether this approach is “flipping the microenvironment” from an immunosuppressive myeloid-dominant one to one in which CD8-positive effector T cells are penetrating. This will be studied using bulk and single-cell RNA sequencing, state-of-the-art immune mass cytometry, and quantitative stromal ­immunohistochemistry. 

DISCLOSURE: Dr. Shafique has served in a consulting or advisory role for GlaxoSmithKline; has received institutional research funding from Amphivena, Bristol-Myers Squibb, Genentech, Merck Serono, Nektar, PsiOxus Therapeutics, and Vaccinex; and has been reimbursed for travel, accommodations, or other expenses by Janssen Research & Development, Nektar, and Vaccinex. Dr. Ruffolo reported no conflicts of interest.

REFERENCES

1. Shafique MR, Fisher TL, Evans EE, et al: Interim results from a phase Ib/II study of pepinemab in combination with avelumab in advanced non-small cell lung cancer patients following progression on prior systemic and/or anti-PDx therapies. 2020 ASCO-SITC Clinical Immuno-Oncology Symposium. Abstract 75. Presented February 6, 2020.

2. Ruffolo LI, Ullman NA, Dale B, et al: Antibody blockade of semaphorin 4D to sensitize pancreatic cancer to immune checkpoint blockade. 2020 ASCO-SITC Clinical Immuno-Oncology Symposium. Abstract 26. Presented February 6, 2020.