Commentary

The introduction of immune checkpoint inhibitors that target the programmed cell death 1 (PD-1)/programmed cell death ligand 1 (PD-L1) pathway has significantly transformed the landscape of cancer therapy since the approval of the first drug in its class, pembrolizumab, in 2014. PD-1 and PD-L1 form an immune checkpoint that is known to suppress the T-cell immune response. In the realm of T-cell biology, it has received particular attention for its role in modulating tumor cell immune tolerance. Tumor cells express PD-L1, which interacts with PD-1 on T cells, inducing apoptotic pathways and downregulating the immune response within the tumor microenvironment.1 Therefore, pharmacologic inhibition of the PD-1/PD-L1 interaction prevents tumor cells from evading the immune system and has demonstrated efficacy in a variety of solid tumors, including but not limited to melanoma, non-small cell lung cancer, urothelial carcinoma, and hepatocellular carcinoma.1

As we eclipse more than a decade since the approval of the first checkpoint inhibitor, we are observing an increasing number of patients who have enjoyed durable responses for tumor types that previously had devastatingly poor outcomes. Over the last decade, long-term survival curves for patients with unresectable advanced melanoma receiving pembrolizumab or ipilimumab from the KEYNOTE-006 trial have been closely monitored. Pembrolizumab’s efficacy in advanced melanoma was confirmed with the following results: (1) the pembrolizumab group had a median melanoma-specific survival of 51.8 months compared with (2) the ipilimumab group’s median melanoma-specific survival of 17.2 months.2 After 10 years, both survival curves exhibited a stable tail end, supporting the idea that these patients may even be cured of their metastatic disease. However, there are subsets of patients whose tumors show resistance patterns and have no clinical response to therapy. Possible mechanisms of resistance include: (1) alternative immune checkpoints that inhibit lymphocyte activity and may upregulate in response to PD-1 suppression, (2) irreversible T-cell exhaustion, (3) insufficient tumor mutational burden to trigger the necessary antigenic activation of cytotoxic lymphocytes, and (4) the presence of alternative immunosuppressive cells in the tumor microenvironment that modulate the PD-1/PD-L1 pathway.3 Evolving research is currently underway into other resistance pathways as our understanding remains incomplete.

In recent years, research has been growing on tumor-derived exosomes (TDEs) and their role in the PD-1/PD-L1 pathway. Tumor cells that express PD-L1 also produce exosomes with PD-L1 on their surface, and a correlation has been found between the presence of PD-L1-expressing TDEs and the development of resistance to anti-PD-1 therapies. In a mouse breast cancer in vitro model, tumor cells pretreated with an exosome secretion inhibitor GW4869 exhibited a significantly improved response to anti-PD-L1 therapy compared with the same tumors without any GW4869 pretreatment.4 Recent studies also indicate that melanoma TDEs express higher levels of PD-L1 than the original melanoma cells from which they were derived, and these TDEs can enter the peripheral vasculature to systemically suppress immune cells.5 Locally, TDEs expressing PD-L1 can also modify the behavior of immune cells within the tumor microenvironment. In glioblastoma multiforme, TDEs have been shown to induce PD-1+non-classical monocytes into immunosuppressive effector cells.6 Presumably, these concepts should be generalizable to different tumor types, as there is nothing uniquely intrinsic about the exosomal biology of the solid tumor models discussed above; these studies highlight a target on TDEs where inhibiting release could enhance response in any tumor type currently treated with anti-PD-1/PD-L1 therapies.

Multiple pathways targeting exosomal PD-L1 have been reported in the literature. Pharmacologic studies have identified potential off-label uses for already Food and Drug Administration (FDA)-approved drugs, including: (1) macitentan, originally developed for pulmonary hypertension, which antagonizes endothelin receptor A implicated in exosome biogenesis, and (2) sulfisoxazole, a sulfonamide antibiotic that also antagonizes the same endothelin receptor A.^{7 8} On the cellular level, activation of endothelin receptor A upregulates the RAB27A and RAB5A genes crucial for extracellular vesicle secretion pathways essential for exosome biogenesis.8 Therefore, an antagonist to endothelin receptor A ultimately suppresses exosome release. Statin therapies have also been researched due to their observed potential antitumor efficacy; it is known that abnormal cholesterol and mevalonate synthesis are associated with tumor growth in breast cancers, as 27-hydroxycholesterol functions as an estrogen receptor modulator.9 The downstream effects of this interaction have been shown to increase exosome secretion, raising the question of whether statin therapies might be effective in preventing exosome biogenesis in breast cancer. Choe et al used a mouse breast cancer model to demonstrate that treatment with atorvastatin suppressed the activity of PD-L1 on TDEs, sensitizing tumor response to subsequent anti-PD-L1 therapy.9

In addition to exploring the anti-exosome activity of existing drugs, there has also been significant interest in developing novel agents specifically aimed at downregulating tumor exosomes. Yang et al tested an exosome secretion inhibitor GW4869 alongside a PD-L1 antibody in a mouse breast cancer model. They confirmed a substantial decrease in exosome secretion, which was coupled with increased sensitivity to the PD-L1 antibody, resulting in enhanced tumor growth suppression.4 Given the notable difference in tumor response with the addition of GW4869 to anti-PD-L1 antibody therapy, it is likely that a component of the PD-L1/PD-1 pathway is being targeted either synergistically or more effectively by the exosome secretion inhibitor than by the anti-PD-L1 antibody. However, further research is needed to elucidate the mechanisms underlying this observed difference. Mechanistically, GW4869 acts as an inhibitor of neutral sphingomyelinase (nSMase), an enzyme that hydrolyzes sphingomyelin to form the lipid raft domains necessary for exosome budding. It is the most widely used exosome secretion inhibitor in translational research, although there are many other sphingomyelinase inhibitors available, such as imipramine, spiroepoxide, and DPTIP (2,6-dimethoxy-4–(5-phenyl-4-thiophen-2-yl-1H-imidazole-2-yl)-phenol). While DPTIP is the most potent drug in its class, GW4869 is more cell-permeable and selective for membrane nSMase, making it the more popular option.10 The minor differences between drugs in the same class underscore the potential for the future development of increasingly optimized exosome inhibitors.

This work has already captured the evolving interest of select pharmaceutical laboratories focused on developing drugs that can modulate exosome activity. To make progress in countering the immune surveillance mechanisms of tumor cells, it is critical to consider the physiological properties of exosomes as independent extracellular vehicles through which tumor cells can regulate immune cells from afar. We aim is to direct the attention of interested research entities toward exosomes as a novel target for immune-checkpoint therapies by highlighting the potential of the preliminary work on this topic. Ultimately, our goal is to advance this work from the preclinical testing phase and begin exploring whether exosome-based therapies could augment existing anti-PD-L1 treatments. To date, trials involving a combination therapy of an exosome secretion inhibitor and an anti-PD-L1 antibody have been conducted only in animal models.4 With current advances in our understanding of exosome biology, we firmly believe that our scientific community is on the verge of initiating Phase 0 and Phase I clinical trials to assess the potential of anti-exosomal-PD-L1 therapies. Several candidate drugs have already been identified in the existing literature: (1) various FDA-approved drugs such as macitentan, sulfisoxazole, and atorvastatin exhibit off-label activity against exosome secretion, and (2) compounds like GW4869 are in the development pipeline as specific and potent exosome secretion inhibitors.4 7–9 Overall, we are optimistic about the potential for clinical advancement in treating any tumor type historically responsive to anti-PD-L1/PD-1 therapies.