The advent of immune checkpoint inhibitors (ICIs) has represented a breakthrough in the treatment of many cancers, although a high number of patients fail to respond to ICIs, which is partially due to the ability of tumor cells to evade immune system surveillance. Non-coding microRNAs (miRNAs) have been shown to modulate the immune evasion of tumor cells, and there is thus growing interest in elucidating whether these miRNAs could be targetable or proposed as novel biomarkers for prognosis and treatment response to ICIs.
1. Introduction
Study of the tumor microenvironment has uncovered a wide battery of mechanisms exploited by neoplastic cells to evade immune system action, which enables them to survive and fosters tumorigenesis
[1]. Immune cell activation requires T cell receptor (TCR) recognition of antigens presented by major histocompatibility complex class I or II molecules (MHC-I/II) expressed on normal cells or antigen presenting cells (APCs), and a costimulatory pathway wherein the receptor CD28 binds to ligands expressed on APC membrane such as B7 family ligands B7-1/CD80 or B7-2/CD86. This interaction stabilizes the signal and triggers the complete activation, proliferation, survival, and cytokine production of T lymphocytes. This same signaling initiates a regulatory mechanism to prevent overactivation of the immune system, in which the expression of immune checkpoint (IC) molecules such as cytotoxic T lymphocyte-associated antigen-4 (CTLA-4; CD152) is upregulated on the lymphocyte cell membrane after T cell activation to compete for the same ligands as CD28, avoiding T cell overactivation and hyperactivity. Similarly, programmed cell death protein 1 (PD-1; CD279) and its ligands PD-L1 (PD-1 ligand 1; CD274; B7-H1) and PD-L2 (CD273; B7-DC); or B7-H3 (CD276) and its ligandstriggering receptor expressed on myeloid cell (TREM)-like transcript 2 (TLT-2), toll-like receptor 2 (TLR2) or interleukin-20 receptor subunit alpha (IL20RA), that inhibit T cell effector functions and induce T cell apoptotic death
[2][3][4][5][6]. Another IC is the cluster of differentiation 47 (CD47), which binds to signal regulatory protein alpha (SIRPα) of membrane macrophages to inhibit tumor cell phagocytosis
[7].
The use of inhibitors against these ICs (immune checkpoint inhibitors (ICIs), also known as immune checkpoint blockade (ICB)) has shown clinical benefits in the treatment of different types of cancer. The inhibitors currently available are PD-1 inhibitors (i.e., Cemiplimab, Nivolumab and Pembrolizumab, and the recently approved Retifanlimab and Dostarlimab, approved in March and July 2023, respectively;
www.fda.gov, accessed on 8 November 2023), and PD-L1 inhibitors (i.e., Atezolizumab, Avelumab and Durvalumab), and CTLA-4 inhibitors (Ipilimumab)), all approved by the U.S. Food and Drug Administration (FDA)
[8]. However, treatment response depends on tumor type; good responses can be observed in immunogenic tumors, as seen in metastatic melanoma, which show a five-year overall survival (OS) rate of up to 52% when combining Nivolumab and Ipilimumab (compared to the five-year OS rate of about 35% for targeted therapy)
[9]. Nonetheless, other tumors do not respond well to ICIs, and these inhibitors can even favor tumor progression, as has been observed in T cell leukemia–lymphoma after treatment with Nivolumab
[10].
This differential response to ICIs highlights the need to find novel biomarkers that can guide decision making to select the most personalized treatment for each patient. The FDA has approved different biomarkers for ICIs, such as PD-L1 expression in tumor cells, microsatellite instability (MSI), and Tumor Mutational Burden (TMB), referring to the totality of somatic mutations (single nucleotide polymorphisms (SNPs) and variations of copy number (CNVs)) per million bases. Other biomarkers are being studied, such as the tumor proportion score (TPS; evaluates expression of PD-L1 on tumor cells)
[11], tumor immune dysfunction and exclusion (TIDE) signature (stratifies patients into high or low cytotoxic T lymphocyte count based on gene signature)
[12], and Immunoscore (based on the density of total and cytotoxic tumor infiltrating T cell). However, these biomarkers present certain limitations of use for selecting the most appropriate individual treatment.
1.1. miRNAs That Modulate Response to ICIs
According to Table 1 and Figure 1, increases in let-7a, let-7b, miR-15b-5p, miR-16-5p, miR-20b-5p, miR-128a, miR-582, miR-708, and miR-4759 levels are positively correlated with increased effectiveness of ICI therapy, while miR-21 and miR-340 presented a reduced response to ICIs in the analyzed models. miR-424 and miR-155 produced opposing outcomes according to the tumor type studied.
Figure 1. Scheme representing relevant miRNAs with the ability of modulating the effectiveness of immune checkpoint inhibitors (ICIs) by directly targeting ICs or other related modulators. Black: high miRNA levels enhanced ICI efficacy; Red: low miRNA levels enhanced ICI efficacy. APC: antigen-presenting cell; BMI1: BMI1 proto-oncogene, polycomb ring finger; CD47: cluster of differentiation 47; CD80: CD80 molecule (B7-1, CD28LG); CTLA-4: cytotoxic T lymphocyte-associated antigen-4 (CD152); p53: tumor protein p53; PD-1: programmed cell death protein 1 (CD279); PD-L1: programmed cell death protein 1 ligand (CD274; B7-H1); PTEN: phosphatase and tensin homolog; SIRPα: signal regulatory protein alpha; STAT1: signal transducer and activator of transcription 1; TCF-4: T Cell Factor 4; TLR2: toll-like receptor 2. Created with Bio.Render.com.
Table 1. miRNAs that modulate response to immune checkpoint inhibitors (ICIs).
miRNA |
Cancer |
miRNA Target Gene |
ICI |
Experimental Model |
Effect on ICI Response |
Refs. |
let-7a and let-7b |
Head and neck squamous cell carcinoma |
TCF-4 * |
anti CTLA-4 |
Overexpressing let-7a/b tumor cells inoculated into mice + anti CTLA-4 |
H |
[13] |
miR-15b-5p |
Colorectal cancer |
PD-L1 * |
anti PD-1 |
Overexpressing miR-15b-5p tumor cells inoculated into mice + anti PD-1. |
H |
[14] |
miR-16-5p |
Lung cancer |
|
anti PD-L1 |
Tumor cell + overexpressing miR-16-5p exosomes + anti PD-L1 |
H |
[15] |
miR-20b-5p |
Lung cancer |
PD-L1 * |
anti PD-1 |
Tumor cells transfected with miRNA mimic + Pembrolizumab |
H |
[16] |
Breast cancer |
Tumor cells transfected with miRNA mimic + Pembrolizumab |
H |
miR-21 |
Oral squamous cell carcinoma |
PTEN |
anti PD-L1 |
Tumor cells inoculated into mice + miR-21 knockdown tumor-derived exosomes + anti-PD-L1 |
L |
[17] |
Melanoma |
STAT1 * |
anti PD-1 |
Tumor cells and knocked down miR-21 tumor-associated macrophages (TAM) subcutaneously injected in mice + anti PD-1 |
L |
[18] |
miR-128a |
Laryngeal squamous cell carcinoma |
BMI1 * |
anti PD-1 |
Overexpressing miR-128a tumor cells + Pembrolizumab |
H |
[19] |
miR-155 |
Metastatic melanoma |
|
anti PD-1 + anti PD-L1 + anti CTLA-4 |
Tumor cells inoculated into modified mice for knockout of miR-155 in CD4/8 T cells + anti PD-1, anti PD-L1 and anti CTLA-4 |
L |
[20] |
Diffuse large B-cell lymphoma |
PD-L1 * |
anti PD-L1 |
Overexpressing miR-155 tumor cells inoculated into mice + anti PD-L1 |
L |
[21] |
Breast cancer |
SOCS1 |
anti PD-L1 |
Overexpressing miR-155 tumor cells inoculated into mice + anti PD-L1 |
H |
[22] |
Melanoma |
PD-L1 * |
anti PD-L1 |
Overexpressing miR-155 tumor cells co-cultured with peripheral blood mononuclear cells + anti PD-L1 |
H |
[23] |
miR-340 |
Pancreatic carcinoma |
CD47 * |
anti CD47 |
Overexpressing miR-340 tumor cells inoculated into mice + anti CD47 |
L |
[24] |
miR-424 |
Colorectal cancer |
CD28 and CD80 * |
anti PD-1 + anti CTLA-4 |
Tumor cells inoculated into miR-424 knocked mice + anti PD-1 and anti CTLA-4 |
L |
[25] |
Mouse cecum orthotopic colorectal cancer + miR-424 knocked tumor cell-derived extracellular vesicles + anti PD-1 and anti CTLA-4 |
L |
Hepatocellular carcinoma |
PD-L1 |
anti PD-L1 |
Tumor cells inoculated into mice + nanobubbles carrying miR-424 mimic and anti PD-L1 |
H |
[26] |
miR-582 |
B-cell precursor acute lymphoblastic leukemia |
CD276 * |
anti CD276 |
Overexpressing miR-582 tumor cells co-cultured with NK cells + anti CD276 |
H |
[27] |
miR-708 |
T-acute lymphoblastic leukemia |
CD47 * |
anti CD47 |
Overexpressing miR-708 tumor cells + anti CD47. |
H |
[28] |
miR-4759 |
Breast cancer |
PD-L1 * |
anti PD-L1 |
Overexpressing miR-4759 tumor cells co-cultured with peripheral blood mononuclear cells + anti PD-L1 |
H |
[29] |
1.2. miRNAs Modulated after Response to ICIs
Other articles were found examining correlations between miRNA levels and therapy response, but these were not included due to the fact that response to ICI was determined via biomarkers like IC expression, the TIDE algorithm, or the proportion of immune cells using CIBERSORT-type algorithms (a deconvolution method that characterizes the cell composition of the sample based on gene expression profiles)
[30]. Other articles excluded were those which identified miRNAs targeting ICs but did not validate the impact of miRNA deregulation on ICI response. Lastly, studies analyzing miRNA levels only before or after treatment (common in biomarker identification studies) were omitted for not experimentally establishing the definitive link between changes in miRNA expression and ICI response.
Table 2 shows miRNAs modulated after response to ICIs in patients. Good responders (complete response, partial response, or stable disease) to ICIs saw increased miR-22, miR-24, miR-99a, miR-194, miR-214, miR-335, miR-339, and miR-708 levels, while the overexpression of miR-4649-3p and miR-615-3p correlated with no response to ICIs.
Table 2. miRNAs modulated after response to immune checkpoint inhibitors (ICIs) in patients.
ICI |
Experimental Model |
miRNA |
Experimental Effect on miRNA |
Refs. |
anti PD-1 |
miRNA analysis in peripheral lymphocytes from 21 good responders (complete response, partial response, or stable disease) with metastatic renal cell carcinoma before and after a 4-weeks period (2 cycles) of nivolumab administration |
miR-99a, miR-708, miR-655, miR-582-3p, miR-492, miR-487a, miR-485-3p, miR-449a, miR-433, miR-431, miR-429, miR-376c, miR-342-5p, miR-340, miR-339-5p, miR-335, miR-324-5p, miR-25, miR-24, miR-22, miR-221, miR-214, miR-194, miR-18b, miR-152, miR-143, miR-100, miR-let-7ev |
High levels of expression in peripheral lymphocytes after treatment compared to before treatment in good responders. |
[31] |
miRNA analysis in peripheral lymphocytes from 17 good long-responders (complete response, partial response or stable disease and progression-free survival (PFS) > 18 months) with metastatic renal cell carcinoma before and after a 4-weeks period (2 cycles) of nivolumab administration |
miR-22, miR-24, miR-99a, miR-194, miR-214, miR-335, miR-339, miR-708 |
High expression levels in peripheral lymphocytes after treatment compared to before treatment in good responders. |
anti CTLA-4 + anti PD-1 |
Plasma from stage IV melanoma non-responders (13 patients), partial response (4 patients) and complete response (5 patients) before and after Ipilimumab and Nivolumab/Pembrolizumab treatment |
miR-4649-3p and miR-615-3p |
Increased levels in post- vs. pre-treatment in non-responders. No changes post- vs. pre-treatment in patients with partial response. Decreased levels post- vs. pre-treatment in patients with complete response. |
[32] |
1.3. miRNAs Related to ICIs Response
Figure 2 shows that only the miR-340 and the miR-708 had been studied in more than one category, “miRNAs that modulate response to ICIs” and “miRNAs modulated after response to ICIs in patients”. Among them, miR-708 was the only one with the same results in the different studies analyzed. While miR-340 was related to reduced response to ICIs in some studies, in others, its expression levels (in peripheral lymphocytes) after treatment with ICIs was associated with a good response.
Figure 2. Venn diagram showing miRNAs that were identified as miRNAs able to modulate responses to ICI, miRNAs which were modulated after response to ICIs in patients, as well as miRNAs that can regulate irAEs.
2. miRNAs Related to Immune Checkpoint Inhibitor Response
2.1. miRNAs That Modulate Response to ICIs
Several miRNAs directly or indirectly regulate PD-L1 expression (Table 1). The let-7 family (let-7a, let-7b, let-7c, let-7d and let-7e) is thought to mediate tumor suppression in cancer by inhibiting tumor cell proliferation, promoting cell death evasion or metastasis [33] but also alterations to immunity. Let-7 was significantly downregulated in tissue from head and neck squamous cell carcinoma patients compared to healthy tissue [13]. Let-7 downregulation has been observed in other types of cancer associated with reduced copy numbers, such as melanoma [34], with an upregulation of LIN28A/LIN28B, which is an RNA binding protein that inhibits Drosha or Dicer binding during let-7 biogenesis (breast cancer) [35], and with DNA hypermethylation (epithelial ovarian cancer) [36].