Investigators are working on a new generation of therapies that activate CD40, an immune checkpoint that helps promote an antitumor response by boosting T-cell–stimulatory signals.
Investigators are working on a new generation of therapies that activate CD40, an immune checkpoint that helps promote an antitumor response by boosting T-cell–stimulatory signals. Although efforts to hit this target have often proved disappointing in clinical trials over the past 2 decades, recent strategies focusing on new drug designs and novel combinations are reviving hopes for success.1-3
Ongoing clinical efforts are directed at exploring combinations that could expand the impact of immunotherapy into immunologically “cold” tumors by exploiting complementary mechanisms of action.
In preliminary clinical trial results, APX005M, a CD40-directed monoclonal antibody (mAb) being developed by Apexigen, showed promise for treating notoriously immunotherapy-resistant pancreatic ductal adenocarcinoma (PDAC) in combination with chemotherapy and with or without the anti-PD–1 mAb nivolumab (Opdivo).4
Other pharmaceutical companies are follow-ing Apexigen’s lead and pursuing similar combinations in PDAC with the goal of providing immunotherapeutic inroads into this challenging tumor type.
At least 10 novel agents are being tested in clinical trials, frequently in combination with chemotherapy or immune checkpoint immunotherapy, according to a search of the ClinicalTrials.gov website (Table).
CD-40 is a member of the tumor necrosis factor receptor (TNFR) superfamily, a group of structurally related receptors and their ligands that provide critical communication signals between various cell types, including immune cells, where they play key roles in innate and adaptive immune responses.1-3,5 (Figure6). The goal of therapeutically targeting CD40 is to activate CD40+ antigen-presenting cells (APCs) to boost the antitumor immune response by stimulating tumor-specific cytotoxic T cells.
CD40 is a transmembrane receptor that is constitutively expressed predominantly on APCs, such as macrophages, dendritic cells (DCs), and B cells, but it is also found on plate-lets and some nonhematopoietic cells.1-3 CD40 expression has also been shown on malignant cells across a variety of cancer types, including bladder cancer, melanoma, lung cancer, colorectal cancer, and several B-cell malignancies, and has frequently been associated with improved outcomes.3
The CD40 ligand, CD40L (also known as CD154), is also a transmembrane protein whose expression is induced on the surface of activated CD4+ T cells, B cells, and natural killer (NK) cells, among others.1-3 Interestingly, this receptor-ligand localization pattern is unique within the TNFR superfamily and reflects the early role of CD40 in the adaptive immune response, as the TNF ligand (not the receptor) is typically expressed on the APC.1
CD40L naturally assembles into trimers at the cell surface, and their binding to CD40 receptor molecules leads to receptor clustering and the formation of a signaling complex that is required for intracellular signal transmission via the recruitment of TNFR-associated factors (TRAFs). Depending on the specific TRAFs and other cytoplasmic adapter proteins that are engaged, CD40 activation can trigger a number of downstream signal-ing cascades, including the MAPK, PI3K, and NFκB pathways.1-3
The physiological outcomes of CD40 path-way activation are varied and complex, but one of its central functions is as a major molecular mediator of the capacity of CD4+ helper T cells to trigger tumor cell–specific cytotoxicity.1-3,7
It does this by licensing DCs to activate CD8+ cytotoxic T cells via upregulation of DC-surface expression of major histocompatibility complex molecules (to facilitate antigen presentation) and other costimulatory molecules, which provide the secondary signal for T-cell activation. CD40 pathway activation also promotes the production of proinflammatory immunostimulatory cytokines that further modulate T-cell activity.1-3,7
In the absence of CD40, unlicensed DCs drive T-cell exhaustion and the generation of immunosuppressive regulatory T cells. CD40 plays similar roles in B- and NK-cell activation; thus, its effects span innate and adaptive immune responses, as well as cellular and humoral immunity.1,7
As an immune costimulatory molecule, CD40 is targeted with agonistic compounds. From a pharmacological perspective, agonists are much more challenging to develop than antagonistic agents. Although the major focus has been on CD40 mAbs, the first drug to enter development in the early 2000s was a recombinant form of CD40L. In a first-in-human study, 32 patients with advanced solid tumors or non-Hodgkin lymphoma (NHL) were treated with trimeric recombinant human CD40L.
There were 2 objective responses, including a durable complete response (CR) in a patient with head and neck squamous cell carcinoma. The maximum-tolerated dose was defined by grade 3/4 elevated transaminase levels,8 which are now known to be a class effect of CD40-targeted therapies.1
The clinical development of agonist CD40 mAbs began a few years later, but they had already demonstrated significant promise in preclinical trials.1,2 Unfortunately, that did not necessarily translate into success in human patients; clinical activity with single-agent CD40 mAbs was modest.1-3
The most extensive clinical experience has been with Roche’s selicrelumab (formerly known as RO7009789 and CP-870,893), with development ongoing. In a first-in-human single-dose study (NCT02225002), selicrelumab produced partial responses (PRs) in 27% of patients with advanced melanoma. One patient continued to receive selicrelumab every 1 or 2 months for a year (NCT02157831) and achieved complete remission, which is ongoing 15 years later without additional therapy. Among patients with other solid tumors, however, the best response to single-dose selicrelumab was stable disease (SD).1,9,10 In a subsequent trial, weekly infusions of selicrelumab produced no objective responses, even among patients with melanoma.11 Development is now moving forward with combinatorial strategies.
Preclinical and clinical studies have revealed new insights into the role of the CD40 pathway in normal physiology and the potential antitumor activity of CD40 agonists; this knowledge is informing ongoing research. In addition to their effects on the immune system, CD40 agonists can directly target CD40-expressing tumor cells and have been shown to directly inhibit tumor cell proliferation and trigger apoptotic cell death.6,12
CD40 also has an important T-cell– independent function in monocyte differentiation, preferentially driving the development of the proinflammatory, tumoricidal (M1) type of macrophages, which could also contribute to the antitumor activity of CD40targeted drugs.1,2
Importantly, because CD40 mAbs are bivalent, CD40 cross-linking facilitated by Fcγ receptors (FcγRs) is thought to be required for their agonist activity, as clustering of the CD40 receptor triggers intracellular signaling. Engagement of FcγRs promotes several effector functions of mAbs, including antibody-dependent cellular cytotoxicity, thus representing an additional mechanism by which some CD40 agonists might mediate target cell elimination.2,13
Although FcγR-mediated cross-linking is required for the mouse mAbs used in preclinical development, ongoing debate exists over its importance for the human mAbs in clinical development.1,14,15
Selicrelumab is an immunoglobulin G2 (IgG2) mAb that has been shown to have very low binding affinity to human FcγRs and to have FcγR-independent agonist activity as a result of structural features.16,17 Other authors, however, have concluded that all human CD40 mAbs require FcγR-mediated cross-linking for optimal activity and that the IgG2 design of selicrelumab is the major impediment to its antitumor efficacy.15 Meanwhile, some investigators argue that limiting FcγR-mediated cross-linking and the associated enhancement of mAb activity could help mitigate the toxicity of CD40 agonists.6
This debate and the search for new CD40targeted drugs that strike an optimal balance between antitumor efficacy and toxicity are the driving forces behind ongoing drug development.
Some drug companies have pursued IgG2 mAbs similar to selicrelumab and attempted to reduce agonist activity, following the logic that the toxicity associated with very potent agonists means they must be administered at low doses. This might not achieve saturation of CD40 on target APCs because of off-target binding to nonimmune cells with systemic administration.6
CDX-1140 (Celldex Therapeutics) is a fully human IgG2 mAb with lower in vitro agonist activity compared with selicrelumab. In a preclinical study, its effects on immunological parameters were not significantly different from those of selicrelumab, and there was no evidence of clinically meaningful toxicity was observed in cynomolgus monkeys.6
Results from a phase 1 clinical trial (NCT03329950) in patients with advanced solid tumors or NHL were presented at the annual meeting of the Society for Immunotherapy of Cancer in 2019. Among 62 patients (38 with activity assessments available), 6 treated with CDX-1140 monotherapy at doses ranging from 0.01 to 3 mg/ kg and 2 participants who received CDX-1140 plus CDX-301, a DC growth factor, experienced immune SD lasting up to 5.4 months. The most common treatment-related adverse events (TRAEs) were arthralgia, nausea, and fatigue; serious TRAEs included fatigue, cytokine release syndrome (CRS), pneumonitis, nausea, and pyrexia.18
Most companies are developing IgG1-based mAbs, and many are modifying the Fc region of the antibody to enhance FcγR binding in a bid to exploit the activity boost conferred by CD40 cross-linking. Additionally, the absence of a core fucose molecule in the Fc region has been shown to increase IgG1 Fc binding affinity to an FcγR that is involved in antibody-dependent cellular cytotoxicity19; thus, strategies to enhance the potency of CD40 mAbs have included the development of nonfucosylated varieties.
Among these is SEA-CD40 (Seattle Genetics), which investigators are evaluating in an ongoing phase 1 study (NCT02376699), and interim results were presented at the American Society for Clinical Oncology Annual Meeting in 2018. At the time of presentation, 48 patients with advanced solid tumors had received a median of 2 cycles of SEA-CD40 monotherapy at doses ranging from 0.6 to 60.0 µg/kg on day 1 (n = 38) or 30 µg/kg on days 1 and 8 (n = 10) intravenously every 3 weeks.
Dose-limiting toxicities (DLTs) consisted of infusion-related reactions (IRRs) in 5 patients, and the most common treatment-emergent AEs were IRRs, chills, fatigue, nausea, vomiting, dyspnea, and headache. One patient with basal cell carcinoma had a PR, and 10 patients achieved SD.20 Despite being an Fc-modified IgG1 mAb, SEA-CD40 is still reportedly at the weaker end of the agonist spectrum.1
At the other end of that spectrum is APX005M, a humanized rabbit IgG1 Fc-engineered mAb. APX005M also binds the CD40L binding site and is the most similar, in terms of its molecular features and pharmacodynamics, to the mouse mAbs used in preclinical development.1
In a first-in-human clinical trial, APX005M was administered every 3 weeks at doses ranging from 0.0001 to 1.0 mg/kg in 30 patients with advanced solid tumors. Five patients experienced prolonged SD. Most AEs were mild to moderate in severity, with a DLT of grade 3 and higher CRS observed at doses of at least 0.6 mg/kg.21
An Fc-engineered version of selicrelumab is also in development. Investigators at The Rockefeller University in New York, New York, introduced point mutations into the selicrelumab Fc domain to increase FcγR binding and agonist activity. In a mouse model, a V11 Fc variant had improved antitumor activity compared with selicrelumab; higher doses led to even greater activity enhancement but also increased toxicity.15
To reduce the toxicity of CD40 mAbs, investigators are exploring alternative dosing strategies. In a mouse model, directly injecting the V11 Fc variant into the tumor site led to a significantly lower tumor burden compared with systemic administration. Moreover, no substantial toxicity was observed, and many mice displayed long-term survival.22 The anti-CD40 agonist, now called 2141-V11, is being evaluated as an intratumoral injection in patients with skin lesions associated with their cancer in a small phase 1 clinical trial (NCT04059588).
Using recombinant CD40L bypasses the issue of FcγR-mediated cross-linking, and several companies are returning to the idea of designing improved versions of this strategy. AstraZeneca is developing MEDI5083, a fusion protein composed of a trivalent but single-chain CD40L receptor–binding domain linked to a human IgG, thereby generating a hexavalent molecule. Such higher-order receptor/ligand multimers have been shown to be superior at transmitting TNFR signals.2
Gene therapy is also being explored, with oncolytic viruses used for tumor-specific introduction of immunostimulatory transgenes. As the virus replicates inside the cancer cells, it produces the transgene- encoded proteins at the tumor site. This technology was described as turning “tumor cells into drug factories” by John Beadle, MD, CEO of PsiOxus, a UK-based company developing NG-350A, which delivers a transgene encoding an agonist CD40 mAb.23 Lokon Pharma is pursuing a similar strategy; LOAd703 carries transgenes expressing CD40L and a second TNF superfamily ligand, 4-1BBL, to activate an antitumor immune response.24,25
Irrespective of mechanism of action, by far the most promising avenue for CD40 agonists is in combination therapy, and this is where most ongoing clinical efforts are focused. Based on preclinical studies that have demonstrated synergistic activity, the major strategies are combining CD40 agonists with chemotherapy, which may increase neoantigen availability; other T-cell–activating drugs, predominantly those inhibiting PD-1 or its ligand; and antiangiogenic drugs.1-3
The rationale for the latter is that, in addition to blocking aberrant angiogenesis and normalizing the tumor vasculature (which can also improve drug access to the tumor), antiangiogenic drugs can help to relieve immunosuppression and boost T-cell infiltration into the tumor. The combination could make immunotherapy more effective in tumors that are typically immunologically cold and therefore unresponsive.26
In several phase 1 clinical trials, selicrelumab was combined with various forms of chemotherapy, yielding objective response rates (ORRs) of 20% to 40%, depending on the specific type of chemotherapy and the cancer type.27-29 Selicrelumab was also tested in combination with the immune checkpoint inhibitor (ICI) tremelimumab, which targets CTLA-4, in 22 patients with immunotherapy-naïve metastatic melanoma, yielding an ORR of 27.3%, including 2 CRs and 4 PRs (NCT01103635). Nine patients displayed long-term survival (≥3 years).30
Preliminary results from an ongoing phase 1/2 trial of intratumoral injection of LOAd703 in combination with nab-paclitaxel (Abraxane) and gemcitabine in patients with PDAC who are not candidates for complete surgical resection were recently presented (NCT02705196).31 Among chemotherapies, nab-paclitaxel has the added advantage of potentially exerting its own immunostimulatory effects.32 At the time of presentation, 13 evaluable patients had been treated at a dose of 5 × 1010 (n = 3), 1 × 1011 (n = 4), or 5 × 1011 virus particles (n = 6). The best response at the lowest dose was SD, but 6 patients receiving the higher doses experienced PRs. TRAEs were transient and largely grade 1 or 2, most commonly fever, chills, nausea, and increased transaminase levels.31
In a mouse model of pancreatic cancer, the most significant antitumor efficacy resulted from combining a CD40 agonist with both chemotherapy (nab-paclitaxel and gemcitabine) and ICIs. This strategy led to complete rejection of established tumors and long-term survival in a significant number of mice.33
These findings prompted a flurry of clinical trials evaluating this type of triplet therapy in patients with PDAC, a notoriously hard-to-treat and immunotherapy-resistant form of cancer, in addition to other cancer types.
Promising preliminary data from an ongoing trial of APX005M in combination with nab-paclitaxel and gemcitabine, with or without nivolumab, in patients with previously untreated PDAC (NCT03214250) were presented at the American Association for Cancer Research meeting in 2019.
Patients were enrolled in 4 cohorts, all of which received gemcitabine plus nab-paclitaxel as well as APX005M. The APX005M dose was 0.1 mg/kg for cohorts B1 and C1 versus 0.3 mg/kg for B2 and C2. Additionally, cohorts C1 and C2 received nivolumab. At the time of presentation, 24 patients were evaluable for DLTs (6 per cohort), and median follow-up was 32.2 weeks. Fourteen patients experienced PRs (11 confirmed), and 8 achieved SD. There were 2 DLTs: grade 3 and 4 febrile neutropenia in the B2 and C1 cohorts, respectively. Ten patients experienced a serious TRAE, and 2 patients died as a result of AEs (sepsis and septic shock in the setting of neutropenia).4
Meanwhile, the ongoing phase 1 study of SEA-CD40 has been expanded to include a new cohort that is testing the combination of SEA-CD40 with gemcitabine, nab-paclitaxel, and pembrolizumab (Keytruda) as a frontline treatment for patients with metastatic PDAC.34