Phosphodiesterases (PDEs) are a class of enzymes that catalyze the cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP) (1). The PDE superfamily is divided into 11 families (PDE1-PDE11), classified by structure and enzymatic properties (2). Serving as a secondary messenger, cAMP plays a crucial role in regulating cell proliferation (3–5), differentiation (6, 7), apoptosis (8–11), and immune responses (9, 12, 13), by influencing the development of both physiological and pathological processes. The PDE4 family is arguably the most studied of all the PDE families. However, as an important PDE subtype that catalyzes the hydrolysis of cAMP, PDE4 has gradually attracted researchers’ attention for its role in modulating a variety of pathological processes. The PDE4 family comprises several subtypes, notably including PDE4A, PDE4B, PDE4C, and PDE4D. These subtypes are respectively located on different chromosomes: PDE4A at 19p13.2, PDE4B at 1p31, PDE4C at 19p13.11, and PDE4D at 5q12 (14). Each subtype can be divided into long, short, and ultrashort forms based on molecular weight. This diversity underpins the specific functions of PDE4 in various cell types and tissues (15). These subtypes exhibit differential cellular distributions and carry out unique roles in both physiological and pathophysiological processes. For example, PDE4B is highly expressed in neutrophils and other immune cells (16–18), closely related to inflammatory responses. Whereas PDE4D is abundant in the heart (19, 20) and brain (21–23), contributing to cardiovascular function and memory formation. The specific expression patterns and functions of PDE4 subtypes offer possibilities for developing targeted therapies for specific diseases.

Growing evidence highlights the complex relationships between PDE4 and various aspects of cancer. By regulating cAMP levels, PDE4 subfamilies were involved in tumorigenesis (24–28), metastasis (29–31), drug resistance (32, 33), and immune microenvironment (34–37). However, due to the PDE4 family having multiple subtypes and the heterogeneity of tumor cells, the role of PDE4 in tumor progression is also complex and varied. For instance, PDE4A appears to act as an oncogene in hepatocellular carcinoma cell line Huh7 (38), whereas it exhibits tumor-suppressive effects in HepG2 and BEL7402 cells (39). Thus, the PDE4/cAMP signaling pathway has emerged as a promising but extremely challenging target for drug development due to its strong association with tumor progression.

Advancements in understanding the structure and function of PDE4, have propelled the development of small molecule inhibitors targeting PDE4. Several PDE4-selective inhibitors have successfully reached the market, targeting diseases such as psoriasis, atopic dermatitis, chronic obstructive pulmonary disease (COPD), asthma and alleviating pain caused by smooth muscle spasm ( Table 1 ). More recent work is examining PDE4 in the context of cancer, including B-cell lymphomas (40), triple negative breast cancer (41), and colorectal cancer (42). These studies offer promising insights into the potential of PDE4 inhibitors to target tumors.

This review aims to provide a comprehensive overview of the latest progress in understanding the relationship between PDE4 and cancer, examining its molecular mechanisms in tumorigenesis, drug resistance, and immunotherapy, summarizing the status and challenges of PDE4 inhibitors in tumor treatment, and prospecting the future research directions.

Various patterns of expression and localization within cells and tissues significantly influence the specific roles and functions of PDE4 subfamilies and isoforms (15). Additionally, researchers have found that the expression of different PDE4 subtypes in tumor cells varies significantly ( Figure 1 ). It suggests that PDE4 could play either a promotional or inhibitory role in the development and progression of tumors, depending on the type of disease and its stage of advancement.

Drug resistance is the main cause of tumor treatment failure and recurrence. The causes of tumor cell resistance are multiple, including reduced intracellular drug accumulation (109), increased drug efflux (109), alterations in DNA repair mechanisms (110), cancer stem cells (CSC) (111), and modifications in cell death pathways (112). PDE4 subtypes are abnormally expressed in cancer stem cells from various tumors, including breast cancer (111), glioma (37, 113), prostate cancer (114), and leukemia (115). For example, the expression of the PDE4A1 protein is increased in breast CSCs with reduced levels of cAMP (111). The combination of rolipram with paclitaxel demonstrated a synergistic effect and effectively inhibiting the growth of CSCs. Tamoxifen is widely used as the standard first-line adjuvant therapy for estrogen receptor (ER)-positive breast cancer patients, particularly in premenopausal women (116). These studies have demonstrated that tamoxifen administered as adjuvant therapy for at least five years reduces the recurrence rate and mortality of breast cancer by roughly half and one-third, respectively (117). Nevertheless, despite its clinical successes, tamoxifen encounters a significant challenge in that approximately 20%-30% of high-risk, late-stage ER-positive breast cancer patients develop either de novo or acquired resistance to the drug (33, 118). The mechanism of tamoxifen resistance is partially understood. By whole-transcriptome sequencing, PDE4D was identified as a potential mediator of tamoxifen resistance (33) ( Figure 5 ). Moreover, an elevation in PDE4D expression has been observed in tamoxifen-resistant cells and breast cancer tissues. Resistance to tamoxifen treatment was reversed by PDE4D siRNA or cAMP analogs, which restore sensitivity by activating JNK and P38 pathways and concurrently suppressing AKT activation. Interestingly, aspirin was identified to influence cAMP levels both in vitro and in vivo by modulating PDE4D activity, and overcoming tamoxifen resistance (33). Acquired resistance to standard-of-care (SOC) treatments for estrogen receptor-positive (ER+) breast cancer, including endocrine therapy and CDK4/6 inhibitors, significantly diminishes patient survival (119, 120). In non-resistant cells, SOC treatment downregulates PDE4D, resulting in elevated cAMP levels, which prompts PKA-dependent phosphorylation of mitochondrial COXIV-I, the generation of reactive oxygen species (ROS), and DNA damage (121). However, during the development of SOC resistance, an ER-to-EGFR switch activates c-Jun, which in turn induces overexpression of PDE4D. Of note, combining SOC with inhibitors of PDE4D, EGFR, or PARP1 has been shown to overcome SOC resistance (121) ( Figure 5 ). As a rare and highly aggressive form, inflammatory breast cancer (IBC) presents distinctive clinical features with a low survival rate (122). At the same time, PDE4A exhibited a special role in paclitaxel-resistant IBC cells (123). An increase in AMP, cAMP, and PDE4A expression is observed in paclitaxel-resistant cells, suggesting a potential upregulation of the entire cAMP pathway (123). However, the knockdown of PDE4A has a minimal impact on paclitaxel sensitivity, which may be attributed to the activation of other PDE4 subtypes. Further investigation revealed that pSTAT3 regulates the expression of genes associated with inflammation, EMT, and PDE4A in resistant cells, contributing to paclitaxel therapeutic resistance.

Studies have found that miR-494 is significantly downregulated in gastric cancer cells compared to normal gastric epithelial cells. In addition, its expression is reduced in doxorubicin-resistant gastric cancer cells (AGS/dox) compared to parental cells (32). The increased expression of miR-494 inhibits the expression of PDE4D mRNA and protein in gastric cancer cells. Luciferase assays show that miR-494 directly targets the 3’ untranslated region (3’UTR) of PDE4D. Additionally, restoration of PDE4D reverse the drug sensitivity of in gastric cancer cells with miR-494 overexpression (32). Studies have confirmed that PDE4D is the predominant subtype of PDE4 in either normal or neoplastic renal cells (108). Knockdown of PDE4D or treatment with PDE4 inhibitors enhances the inhibition of sorafenib (108). Similarly, PDE4 inhibitors have also promising synergistic anti-tumor effects in B-cell lymphoma (124, 125).

Cancer immunotherapy represents a revolutionary approach in cancer (126, 127). However, despite remarkable progress and groundbreaking successes, especially with checkpoint inhibitors (128–130) and CAR T-cell therapies (131–133), not all patients respond equally, and many patients become resistant to them. PD-L1 (Programmed Death-Ligand 1) serves as a target for immune checkpoint inhibition in cancer therapy, preventing cancer cells from evading immune system attacks by binding to the PD-1 receptor on T cells. Given the role of PD-L1 in suppressing anti-tumor T cell activity, considerable efforts have been made to identify regulators of PD-L1 expression. In immune system cells, the PDE4 family is responsible for a major portion of cAMP hydrolysis (134). In DLBCL cells, the cAMP effectors, specifically PKA and CREB, stimulate the transcription and secretion of IL-10, IL-8, and IL-6. This initiates an autocrine feedback loop, which in turn activates the JAK/STAT signaling pathway, ultimately resulting in elevated expression of PD-L1 on the cell surface (35). Among the members of the PDE4 family, PDE4B stands out as a pivotal regulator of cAMP levels in lymphocytes (135, 136). In PDE4B knockout mice, the proportion of B cells and T cells expressing PD-L1 is significantly higher than in wild-type mice (35). Crucially, despite its broad immunosuppressive properties, the PDE4 inhibitor roflumilast did not diminish the clinical activity of checkpoint inhibitors in a B-cell lymphoma mouse model. Indeed, PDE4 inhibition led to the emergence of a favorable anti-tumor immune profile, including a relative increase in CD8+ cytotoxic T cells (35). Similarly, the PDE4/cAMP pathway may also play a crucial role in regulating PD-L1 and immune infiltration in LUAD (137). Tumor mutation burden (TMB), defined as the number of somatic mutations in a tumor genome after excluding germline mutations, is widely regarded as a potential predictor of immune responsiveness. In a study, bioinformatic screening of LUAD cases with varying TMB values identified PDE4D as a gene associated with the efficacy of immunotherapy (34). High expression of PDE4D correlates with reduced infiltration of CD8+ T cells, suggesting a relative inhibition of the tumor immune microenvironment in LUAD. Moreover, increased PDE4D expression in LUAD patients who were insensitive to PD-1 therapy was consistent with reduced infiltration of CD8+ T cells. Mechanistically, PDE4D negatively regulates the expression and secretion of IL-23 in LUAD cells through cAMP. IL-23 derived from tumor cells fosters a self-amplifying loop of IL-9 secretion through an NF-κB-dependent mechanism, thereby enhancing the cytotoxic function of cytotoxic T lymphocytes (34) ( Figure 5 ). These research findings may help to inform the clinical application of FDA-approved PDE4 inhibitors.

Glioblastoma (GBM) is often referred to as “immune cold” characterized by low T cell infiltration. This is not only attributed to the low expression of tumor antigens and a highly immunosuppressive tumor microenvironment but also to the tightly regulated blood-brain barrier that hinders the infiltration of immune cells into the tumor (138). Endothelial cells derived from glioblastomas (GdECs) are thought to arise from glioblastoma stem cells (GSCs) and participate in the formation of vasculogenic mimicry, which facilitates the growth and invasion of cancer (139). Through a rigorous drug screening approach, two cAMP activators have been identified as effectively inhibiting tubule structure formation in human GSCs (140). Indeed, combining cAMP activators with PD-1 blockade therapy resulted in a significant extension of survival in mice, accompanied by the observation of a robust memory immune response (141). Unstable responses to bevacizumab pose a major challenge in the antiangiogenic treatment of high-grade gliomas, which appears to be associated with increased levels of HIF1α and activation of AKT within tumors following treatment (142, 143). Studies show that there is an interaction between PDE4A and HIF1α in tumor cells (58). Selectively inhibiting PDE4 not only blocks this interaction but also suppresses angiogenesis and enhances the efficacy of bevacizumab in an orthotopic glioma model, opening new directions for anti-angiogenic treatment in malignant gliomas (36).

Rolipram and Ro 20-1724 were first identified as high-selectivity PDE4 inhibitors (144). Subsequently, many specific PDE4 inhibitors have been found and are considered anti-inflammatory agents for asthma and COPD, as they can reduce oxidative stress, TNF-α production, and cytokine generation. Roflumilast, marketed as Daxas^®, was the first PDE4 inhibitor developed for the treatment of COPD (145). Additional indications are also being explored for roflumilast. Phase IV studies have shown that roflumilast reduce fat mass, leading to weight loss in women with obese polycystic ovary syndrome (PCOS); However, these reductions are less pronounced than those induced by liraglutide (146, 147). Roflumilast has also been tested for its potential to enhance cognition and information processing abilities in healthy individuals, with promising results observed (148). Apremilast, a third-generation PDE4 inhibitor, was approved in 2014 for the treatment of adult patients with psoriatic arthritis (PsA) and moderate-to-severe plaque psoriasis. A significant amount of preclinical data has validated the potential of PDE4 inhibitors as therapeutic agents in the treatment of schizophrenia and cognitive dysfunctions (149–152). Based on preclinical findings, the novel PDE4 inhibitor ASP9831 was tested in Phase I and Phase II trials for nonalcoholic steatohepatitis (NASH), but failed to improve biochemical biomarkers of the disease (153).

Furthermore, a multitude of preclinical and clinical investigations has substantiated the critical role of PDE4 inhibitors in cancer treatments ( Table 2 ). Preclinical studies and clinical data collectively indicate that roflumilast demonstrate antitumor activity in B cell lymphomas (40, 125). In conjunction with cAMP-enhancing agents, the PDE4 inhibitor rolipram has exhibited the ability to inhibit triple-negative breast cancer in vitro and vivo (41). Likewise, apremilast prompted tumor regression in murine models of colorectal cancer (42). Possibly even more significantly, precise inhibition of PDE4D, whether through genetic approaches or by employing the PDE4D inhibitor Gebr-7b, re-established chemosensitivity in estrogen receptor-positive breast cancer cells that had become resistant to chemotherapy (33). Furthermore, studies have shown that rolipram enhance the therapeutic effect of bevacizumab on gliomas in vitro ( 36, 37). A clinical case report describes the use of apremilast in managing pustular psoriasis that occurred concurrently in a patient with Ewing’s sarcoma, following chemotherapy with Ifosfamide and Etoposide (154). This suggests that apremilast may offer a novel therapeutic strategy, particularly in the management of cutaneous side effects associated with chemotherapy.

In pharmacotherapy, a critical challenge lies in determining the minimum effective dose, which ensures therapeutic efficacy while minimizing drug utilization as much as possible. In other words, a central issue in drug development and clinical practice is how to administer the smallest dose of medication while maintaining its therapeutic effect. It was observed that in MCF-7 cells, which are responsive to hormonal stimuli, chylomicrons demonstrated superior performance over liposomes. Additionally, when rolipram was encapsulated within these vesicles, its IC50 value decreased fourfold relative to the administration of free rolipram (165). Conversely, the IC50 value for Rolipram encapsulated in liposomes was notably higher (165). The underlying mechanisms behind these observations can potentially be better understood through detailed molecular docking investigation. This provides significant guidance for achieving more effective drug delivery.

However, the clinical use of PDE4 inhibitors has faced challenges due to their associated side effects. For instance, although rolipram holds potential pharmacological effects, it has a considerably narrow therapeutic window. In clinical trials, adverse reactions such as nausea, vomiting, and headaches frequently occurred. These adverse effects of rolipram have significantly limited its clinical application (174). While roflumilast demonstrated superior performance to rolipram in clinical trials, gastrointestinal adverse reactions still occurred in 9.5% of cases, characterized by symptoms such as diarrhea, nausea, headache, weight loss, urinary tract infections, and psychiatric disorders (175). Nevertheless, patients may experience adverse reactions such as weight loss (10%), nausea (8.9%), diarrhea (7.7%), nasopharyngitis (2.6%), and upper respiratory tract infection (3.9%) (175). Several hypotheses have been proposed regarding the mechanisms behind PDE4-induced side effects (such as hypothermia, dizziness, diarrhea, nausea, vomiting, etc.), including subclass selectivity, partial inhibition, subcellular compartmentalization, and tissue-specific distribution, but none of these hypotheses has been fully substantiated (176). To overcome the adverse effects induced by PDE4 inhibitors, there is a need for more refined target specificity and a deeper exploration of PDE4 subtype particularities. This involves identifying and leveraging those subtypes that do not contribute to adverse reactions as new targets for drug development. It will require researchers to intricately decipher the functional differences between subtypes at the molecular level and to develop next-generation PDE4 inhibitors that can precisely act on the subtypes relevant to therapy without affecting others that may be implicated in side effects.

Emerging evidences shed light on the role of PDE4 and PDE4 inhibitors in cancer. However, the exact function of PDE4 isoforms and inhibitors in tumor development, drug resistance and immunotherapy remain debatable. Because of the ubiquitous distribution of PDE4, this superfamily of enzymes plays a role in a wide variety of biological processes. Nevertheless, the biological roles played by specific PDE4 isozymes vary due to their unique expression patterns at the level of tissues/organs, cell types and subcellular compartments. PDE4 are not only responsible for regulating the total cellular levels of cyclic nucleotides; instead, they generate distinct pockets or nanodomains of cyclic nucleotide signaling. This subcellular compartmentalization of cyclic nucleotide signaling allows a single cell to selectively respond to various extracellular and intracellular signals (15). In addition, different PDE4 inhibitors have different inhibitory effects on PDE4 subtypes, which leads to different effects of PDE4 inhibitors in the treatment of different tumors. For example, zardaverine has shown inhibitory effects on HCC both in vivo and in vitro (177). However, the inhibitory effect of zardaverine on cells appears to be cell-specific, as zardaverine is just not sensitive to cells, including HepG2 cells. More interestingly, the PDE4 inhibitor rolipram failed to mimic this effect, although it also elevated intracellular cAMP levels. Notably, the PDE4 inhibitor rolipram did not have a significant growth inhibitory effect on the zardaverine-sensitive HCC cell line SMMC-7721, which has a relatively low level of PDE4D expression. Consistent with our findings (105), roflumilast inhibited the growth of HepG2 that highly expressed PDE4D. Thus, these HCC cell lines may have different levels of PDE4D expression, and PDE4D may be responsible for cell proliferation during HCC pathology. Again, a similar situation may exist in other tumors. Alternatively, the regulation of the cAMP pathway by PDE4 inhibitors may depend on cell type and conditions. PDE4 inhibitors have the ability to induce the expression of PDE4 isoforms within endothelial and human airway epithelial cells (178, 179). These findings reinforce the notion that specific PDE4 isoforms are segregated by distinct signalosome complexes that regulate local cAMP signaling and confer distinct functional roles.

In addition, the effect of PDE4 inhibitors on tumor still needs more research to confirm. In the future, therapeutic strategies involving PDE4 inhibitors will place a higher value on subtype selectivity and tissue specificity to maximize anti-tumor effects while minimizing side effects. Identifying the patient populations most likely to benefit from PDE4-targeted therapy through precision medicine approaches is critical to improving treatment outcome. Furthermore, combining PDE4 inhibitors with chemotherapy, targeted therapy or immunotherapy will be the focus of research to inhibit tumor growth and restore and enhance the body’s anti-tumor immune response via multiple pathways.

As research progresses and technology advances, our understanding of the functional roles and molecular interactions of each PDE4 isoform and how their function and/or localization is altered in specific diseases grows, it will become more promising to see how precisely we can target PDE4 isoforms to achieve therapeutic efficacy while minimizing side effects. Also, the refinement of these studies is an important basis for the use of PDE4 as a therapeutic target for tumors.