FOXM1: A small fox that makes more tracks for cancer progression and metastasis

Abstract

Transcription factors (TFs) are indispensable for the modulation of various signaling pathways associated with normal cell homeostasis and disease conditions. Among cancer-related TFs, FOXM1 is a critical molecule that regulates multiple aspects of cancer cells, including growth, metastasis, recurrence, and stem cell features. FOXM1 also impacts the outcomes of targeted therapies, chemotherapies, and immune checkpoint inhibitors (ICIs) in various cancer types. Recent advances in cancer research strengthen the cancer-specific role of FOXM1, providing a rationale to target FOXM1 for developing targeted therapies. This review compiles the recent studies describing the pivotal role of FOXM1 in promoting metastasis of various cancer types. It also implicates the contribution of FOXM1 in the modulation of chemotherapeutic resistance, antitumor immune response/immunotherapies, and the potential of small molecule inhibitors of FOXM1.

1. Introduction

Forkhead Homeobox proteins, commonly known as Forkhead Box proteins (FOX), belong to a specialized group of transcription factors (TFs) known as the winged helix TFs, which are evolutionarily conserved and have originated from helix-turn-helix motifs of bacteria [1]. These TFs were first discovered in the eukaryotic Forkhead homeotic gene and eventually brought many significant findings [1]. The mutations of homeotic gene in Drosophila led to a head structure containing two spikes, which is the reason for the name Forkhead Homeobox [1, 2]. These TFs have structural variations around the DNA binding domain, the primary factor responsible for conferring the DNA binding specificities among winged helix proteins [3].

Members of the FOX family proteins have been found to play diverse roles in cell development, maintenance of cardiac equilibrium, angiogenesis, and tumorigenesis of various cancer types [4–10]. Based on sequence conservation, human FOX group proteins were further classified into 19 subclasses (FOX-A to FOX-S) [11, 12]. Further, each subclass have different members, FOX-A (FOXA1, FOXA2, and FOXA3), FOX-B (FOXB1 and FOXB2), FOX-C (FOXC1 and FOXC2), FOX-D (FOXD1, FOXD2, FOXD3, and FOXD4), FOX-E (FOXE1, FOXE2, and FOXE3), FOX-F (FOXF1 and FOXF2), FOXG1, FOXH1, FOX-I (FOXI1, FOXI2, and FOXI3), FOX-J (FOXJ1, FOXJ2, and FOXJ3), FOX-K (FOXK1 and FOXK2), FOX-L (FOXL1, FOXL2, FOXL3, FOXL4, FOXL5, and FOXL6), FOXM1, FOX-N (FOXN1, FOXN2, and FOXN3), FOX-O (FOXO1, FOXO2, FOXO3, FOXO4, FOXO5, and FOXO6), FOX-P (FOXP1, FOXP2, FOXP3, and FOXP4), FOXQ1, FOX-R (FOXR1 and FOXR2), and FOXS1 [11, 13]. Among all FOX family members, FOXM1 is a central TF that plays a critical role in embryonic development, and its aberrant expression is associated with initiation and progression of various cancer types [14–20]. FOXM1 is a crucial regulator of cell proliferation and upregulates the expression of many of the genes associated with DNA replication, G1/S and G2/M transition, cyclin A2, Cdc25A phosphatase, ATF2, and JNK1 [21–25]. It also regulates the expression of several genes necessary for the segregation of chromosomes and cytokinesis at the end of mitosis, such as polo-like kinase/PLK-1, survivin, Aurora B Kinase, CENP isoforms (A, B, and F) [26, 27].

FOXM1 overexpression and/or gene amplification has been associated with various cancers, including prostate cancer (PCa), pancreatic cancer (PC), breast cancer (BC), small cell lung cancer (SCLC), and non-small cell lung cancer (NSCLC) and has been found to be associated with poor prognosis of the diseases [28–34]. Further studies of FOXM1 have revealed its more insidious roles in invasion, migration, angiogenesis, stemness, and therapy resistance [35–40]. A recent interesting study demonstrated that HIF1α transactivates SCL/TAL1-interrupting locus (STIL) protein, which assists FOXM1 in promoting metastasis and stemness of lung cancer (LC) cells by enhancing the expression of downstream targets of FOXM1 [41]. Recent studies have highlighted the prominent role of FOXM1 in facilitating the invasiveness, migration, and distant metastasis of cancer to various organs, along with impairing the prognosis of the disease [30, 42–45]. Here, we envision and compile the outcomes of accumulating studies on the diverse role of FOXM1 in promoting the metastasis of multiple cancers. This review also focuses on the contribution of FOXM1 towards the modulation of chemotherapeutic resistance and the antitumor immune response of current immunotherapies.

2. FOXM1 Structure

TFs of FOX family proteins are evolutionarily conserved, and members of the family have characteristic DNA-binding domain known as forkhead box or winged helix domain [20, 46]. The conserved winged-helix DNA-binding domain contains 90 to 100 amino acids that help in binding to the promoter regions of the gene (Fig. 1A–C). In general, FOXM1 comprises 10 exons, and two of these exons undergo alternative splicing and generate FOXM1 variants with either exon Va or VIIa or both [47]. Due to alternative splicing, FOXM1 has four splicing isoforms named FOXM1A, FOXM1B, FOXM1C, and FOXM1D. The FOXM1A variant is constituted by both exon Va and VIIa [28]. It has no transactivation activity and serves as a transcriptional repressor [28]. The FOXM1D variant is known to promote metastasis. In colorectal cancer (CRC), it has been found that FOXM1D facilitates epithelial-to-mesenchymal transition (EMT) and promotes metastasis using ROCKs activation [48]. The FOXM1C variant functions as a transcriptional activator and is overexpressed in multiple tumors [28]. It does not contain a VIIa exon but has a Va exon in its structure that contains ERK1/2 targeting sequences [49]. FOXM1B lacks exon Va and is the predominantly overexpressed isoform in human cancers [28, 50].

Figure 1: The structural organization of FOXM1 DNA binding domains and mechanisms regulating FOXM1.

(A) Domain wise arrangement for different splicing isoforms of FOXM1. (B) Three-dimensional structural organization of FOXM1 winged-helix-DNA binding domain interacting with the gene promoter. (C) Amino acids residues of FOXM1 showing binding with the nucleotides of DNA helix domain. (D) Cartoon (FOXM1 DNA-binding domain) and ball-stick model (DNA-binding amino acid residues) that could be targeted or utilized for developing small molecule inhibitors to block FOXM1: DNA interactions/binding. The structure coordinates were taken from PDBID:7FJ2. (E) Overview of major signaling pathways regulating FOXM1 expression and inhibition.

Interestingly, the high-resolution three-dimensional structure of human FOXM1 has been solved, providing the structural basis of FOXM1-DNA binding domains (Fig. 1A–C, PDBID: 6OSW; 7FJ2; 3G73) [51–53]. Structurally, FOXM1 comprises three domains, which include a DNA binding domain (DBD), an N-terminal repressor domain (NRD), and a C-terminal transactivation domain (TAD). The interaction of TAD and NRD domains suppress the transactivation of FOXM1 [54]. Interestingly, the human/murine FOXM1 variants show similar specificity for DNA-binding with 5′-A-C/T-AAA-C/T-AA-3′ consensus sequences [19]. The similarities in mice and human FOXM1 DNA binding sites provide an important platform that can be exploited to gain insight into FOXM1 activity in humans. Of interest, Gli1 directly regulates FOXM1 activity in murine stem cells [55]. Similarly, in human colorectal cancer (CRC) and basal cell carcinomas, Gli1 has been reported to regulate FOXM1 [55, 56]. This might have occurred due to the evolutionarily conserved nature of the DNA-binding domains of both human and murine FOXM1. Moreover, this indicates that the target genes for both human and murine FOXM1 might be the same. As a result, investigating murine FOXM1 may allow us to decipher abnormal human FOXM1 activity in various human cancers. Thus, murine models are vital experimental models to elucidate FOXM1 biology.

Forkhead box family of TFs plays a wide range of functions, including cell proliferation, metabolism, apoptosis, and immunity. Based on sequence similarities, nearly 50 FOX proteins have been identified in the human genome, which are further divided into 19 subfamilies. The subfamilies are named alphabetically, from FOXA to FOXS. The FOXM1 subfamily has four known members: FOXM1A, FOXM1B, FOXM1C, and FOXM1D (Fig. 1A). FOXM1 has a significant role in cell-cycle regulation as it controls the transcription of multiple cell-cycle regulators, including, cyclin B2, Cdc25c, and polo-like kinase 1 (PLK1) [57, 58]. Upregulation of FOXM1 has been detected in many cancers including, ovarian cancer (OC), lung cancer (LC), prostate cancer (PCa), pancreatic cancer (PC), breast cancer (BC), and gastric cancer (GC) [59–64]. Interestingly, recent studies have shown that FOXM1 regulates tumor progression, metastasis, and chemoresistance of various cancers (Fig. 1E & Fig. 2) [65, 66]. The involvement of FOXM1 in the progression and metastasis of multiple cancers has accelerated the interest of researchers to utilize inhibitors of FOXM1 as a potential therapeutic approach (Fig. 1E). Owing to this, multiple attempts have been made to develop FOXM1-specific inhibitors; one of the recent specific inhibitors for FOXM1 is FDI-6, which inhibits FOXM1 by interacting with the DNA-binding domain of the protein [67].

Figure 2: FOXM1-associated pathways involved in metastasis, cancer growth and chemoresistance.

(A) FOXM1-mediated metastatic signaling pathways. The estrogen receptor (ER) or human epidermal growth factor receptor 2 (HER2), CXCR4, and TGFβR activates FOXM1 through various pathways such as ERK1/2, SMAD, STAT, and PI3K. Inactivation or deletion of p53/Rb1 also activates FOXM1. The activated FOXM1 translocate to the nucleus and increased the expression of genes associated with epithelial-to-mesenchymal transition (EMT) and angiogenesis such as Snail, Slug, MMPs, VEGF, Zeb-1 that enhances metastasis. (B) FOXM1 upregulation increases cancer cell survival and growth through PLK1, AURKB, DLX1, and CDKs. (C) Modulation of gene expression by FOXM1 infers drug-/chemo-resistance and alteration of immune landscape. FOXM1 activation increases the expression of genes associated with stem cell maintenance such as SOX2, stathmin, UHRF1. It also increased the expression of ABC transporters and PD-L1, the major regulators for chemo-/immuno-therapeutic antitumor response.

2.1. FOXM1 binding motifs for gene regulation

DIV2 motif is identified as a potent binding site for FOXO1 [68, 69]. It has also been found that several FOX TFs showed affinity to this site in their dimeric form, including FOXM1, FOXI1, and FOXO1 [70]. DIV2 site is a tandem DNA motif, which is found in the promoter region of many FOXM1 regulated genes. FOXM1 has a much higher affinity to bind to this site over other DNA binding sites. FOXM1 binding to DIV2 site has a higher thermal stability compared to its binding to other DNA segments and this was observed in electrophoretic mobility shift assays [52]. Interestingly, it was observed that the binding of FOXM1 monomer to DIV2 motif enhances the tendency of the DNA binding domain of FOXM1 to form a dimeric structure, and no monomeric form was observed even at very low protein to DNA molar ratio [52]. Moreover, this binding of FOXM1 to DIV2 site is cooperative in nature. Bioinformatic analysis on MCF-7 cell line showed that 46.3% of total FOXM1 DNA binding occurs at the DIV2 site [70]. Similar observations have been reported in U20S and MDA-MB-231 cell lines through FOXM1-ChIPSeq studies. It was found that FOXM1 binds to the DIV2 motif with a frequency of 22.7% and 22.9% in MDA-MB-231 and U2OS cell lines, respectively [71, 72]. The outcomes of these studies suggest that DIV2 motif may serve as an important regulatory motif in FOXM1 [52].

Interestingly, FOXM1 also binds to the forkhead consensus site/FKH motif, which has a sequence of RYAAAYA (where R is A/G and Y is C/T), with comparatively poor affinity, due to its wing regions having poor DNA contact [53]. It was reported that tandem NTAAACA sequences are preferred by FOXM1, where N can be any base [73, 74]. The investigations concerning affinity and transcriptional activity showed that FOXM1 does not have a high affinity or transcriptional activity over a single site. The mechanism(s) that can describe the preference of FOXM1 to its tandem binding sites have not yet been discovered. Recently, Zhang et al. have shown that the DIV2 site consists of two inversely arranged GTAAACA motifs that may serve as a possible and potent FOXM1 binding site [70]. It has been shown that the promoter sequence of genes targeted by FOXM1 are rich in the DIV2 motif. Based on the widespread presence of tandem DIV2 and DIV0 motifs in FOXM1 targeted promoters, researchers reached the consensus that FOXM1 prefers tandem-repeat DNA binding sites, where they bind cooperatively with the DNA (Fig. 1B–D). This characteristic binding profile may also confer specificity to FOXM1 in the regulation of transcription [59].

3. Role of FOXM1 in cancer progression and metastasis

3.1. FOXM1 in colon cancer

Colorectal cancer (CRC) is one of the major cancer types, and causes the fourth-highest number of cancer-associated deaths yearly [75, 76]. For patients with advanced CRC, the 5-year survival rate is still lower than 15% [77]. It has been found that around 20% of CRC patients have distant metastases at the time of diagnosis. Nearly half of CRC patients develop colorectal metastases, and more than 80% have unresectable liver metastasis. To predict the prognosis of the disease and therapy response, the status of BRAF and KRAS mutations are being routinely analyzed. Another independent prognostic marker for disease-free survival in CRC patients receiving adjuvant chemotherapy would be the status of DNA mismatch repair. Finding new prognostic markers has been essential for therapy, and FOXM1 has been identified as one of those possible markers in many cancers.

To check whether FOXM1 serves as a prognostic marker in CRC, 203 primary colon tumor samples and their adjacent normal colon mucosa and 66 metastatic lymph node samples were analyzed for the FOXM1 expression [75]. Intense FOXM1 staining was observed in the tumor cells compared to their adjacent normal tissue, and interestingly, the metastatic lymph nodes showed even higher FOXM1 expression compared to primary colon tumors. These outcomes clearly suggest that FOXM1 expression significantly correlates with colon cancer disease stage, invasion, as well as distant metastasis [75]. Furthermore, this study demonstrated that high FOXM1 expression is associated with drastically reduced 5-year survival of patients compared to FOXM1 low/negative tumors. They further checked the stage-specific FOXM1 expression status in patient tumor tissues with different stages of CRC patients who underwent radical colectomy. The patients who developed metastasis following the surgery showed significant differences in FOXM1 expression. The patients with high FOXM1 expression have a nearly fourfold high tendency to develop distant metastasis compared to FOXM1 low or negative patients. Overall survival (OS) as well as metastasis free survival (MFS) both were impacted by FOXM1 expression. These findings demonstrated that FOXM1 is directly associated with CRC metastasis.

FOXM1 is also known to interact with PLAUR [75]. PLAUR is the receptor for PLAU, a urokinase-type plasminogen activator. It is known to facilitate invasion and metastasis of tumor cells by degrading components of extracellular matrix (ECM). This is done by converting inactive plasminogen to its active form plasmin mediated by PLAU and PLAUR binding via activating subsequent signaling cascade. The histological analysis of PLAU expression on CRC TMA specimens from the same patient cohort demonstrated that PLAUR expression increased from normal to primary colon tissues, and the highest expression was noticed in metastatic lymph nodes [75]. Interestingly, the PLAUR expression is positively correlated with higher stages (metastatic) of CRC and associated with poor OS and MFS (metastasis free survival). Li et al. observed a positive correlation in the expression of FOXM1 and PLAUR in CRC tissues [75]. The FOXM1 overexpression substantially increased the migration, proliferation, and invasion of CRC cells as suggested by in vitro and in vivo mouse models studies, whereas FOXM1 knockdown reduces cell proliferation, migration, and invasion of CRC cells [75]. The CHIP-coupled promoter studies demonstrated that FOXM1 directly targets the PLAUR promoter (pPLAUR627) in the SW620 cell line. This was further validated through FOXM1 knockdown and mutational studies in the PLAUR promoter (pPLAUR627) [75]. The outcomes of these studies established that the FOXM1-PLAUR axis plays a pivotal role in the progression and metastasis of human CRC.

FOXM1 expression was also found to be positively correlated with pituitary tumor-transforming Gene 1 (PTTG1) expression in CRC cell lines and microarray data [78]. FOXM1 overexpression in SW620 and HCT116 cell lines enhanced the expression of PTTG1, indicating that PTTG1 is regulated by FOXM1. It was demonstrated that FOXM1 transcriptionally activates the expression of PTTG1 by directly binding to the promoter region of PTTG1 [78]. Further, it has been shown that the FOXM1-PTTG1 axis modulates Wnt-signaling by downregulating DKK1. The outcomes of the study suggested that attenuation of FOXM1 and PTTG1 decreased the invasion, migration, and metastasis of CRC cells in vitro and in vivo [78].

3.2. FOXM1 in lung cancer

3.2.1. Non-small cell lung cancer

Lung cancer (LC) is the most frequent form of cancer and one of the leading causes of the highest number of cancer-related mortality worldwide. LC comprises two major histological categories: non-small cell lung cancer (NSCLC) and small-cell lung cancer (SCLC). Amongst all the diagnosed LC patients, 80–85% of patients have NSCLC, with lung adenocarcinoma (LUAD) being the most frequent subtype of NSCLC [79]. At the time of diagnosis, most LC patients have advanced-stage disease accompanied by local or distant metastasis [80]. Based on the prominent molecular targets such as KRAS mutations (G12C/D), EGFR mutations, BRAF, and ALK mutations [81, 82], some of targeted therapies have been developed that benefited LUAD patients [83, 84]. However, due to tumor heterogeneity and lack of universal molecular drug targets, nearly 40–50% of patients did not respond to targeted therapies. Therefore, it is highly desired to uncover the underlying molecular mechanisms of LC and its pathogenesis that will pave the way to develop targeted therapies.

Aberrant expression of FOXM1 has been reported in NSCLC and SCLC, suggesting the crucial role of FOXM1 in lung carcinogenesis [4, 85]. SNAIL, a major regulator of epithelial-mesenchymal transition (EMT), is overexpressed in NSCLC patients and found to be positively correlated with a more aggressive NSCLC phenotype, having a higher risk of distant metastases and lower survival compared to the low SNAIL expression group [86]. Interestingly, the direct association of FOXM1 and SNAIL has been reported in NSCLC, and it was found that FOXM1 directly upregulates the expression of SNAIL through promoter region binding, thus promoting tumor cell invasion and metastasis [87].

It has been further shown that FOXM1 expression was high in stage I and III LUAD patient samples compared to adjacent normal tissues, which suggests that FOXM1 expression correlates positively with disease initiation and progression [87]. The outcomes of the study indicated that FOXM1 could be used as a valuable biomarker for the diagnosis of LC. Following surgical removal of the primary lung tumors, the risk of developing distant metastasis was high in FOXM1^High patients, and the overall survival (OS) of patients was found to be high in FOXM1^Low/negative group. A strong positive correlation was demonstrated between FOXM1 expression and LUAD development and metastasis. FOXM1 overexpression modulates EMT markers, including E-cadherin, vimentin, N-cadherin, and fibronectin, and enhances the invasive and migratory properties of LUAD cells in vitro and in vivo mice models [87]. The outcomes of these studies indicate that FOXM1 transcriptionally activates SNAIL expression, enhancing the invasive and metastatic potential of LUAD cells. The direct association of FOXM1 with SNAIL provides a potential axis for developing future targeted therapies for invasive and metastatic LC.

3.2.2. Small cell lung Cancer

Small cell lung cancer (SCLC) is another isotype of lung cancer and is one of the most aggressive forms of solid cancer. Though it develops as a solid tumor, the burden of circulating tumor cells (CTCs) is incomparable to other solid tumors, making SCLC a systemic disease [88, 89]. Most SCLC patients show distant metastasis at the time of diagnosis [90, 91]. However, little is known about SCLC biology, and that is why there has been an increase in interest among researchers to gain insight into SCLC biology. Liang et al. reported FOXM1 overexpression in SCLC cell lines and patient tumor samples [92]. The transcriptomic analysis demonstrates a positive correlation between the pathways regulating SCLC cell proliferation and DNA damage repair with FOXM1 expression. FOXM1 downregulation sensitizes SCLC cells to chemotherapy. It has been shown that FOXM1 deletion reduced the migration of NCI-H1688 cells in vitro and enhanced the sensitivity of these cells to first-line chemotherapy (cisplatin and etoposide) [92]. They discovered that FOXM1^High is correlated with advanced stage disease, extra-thoracic metastases, and reduced OS of these patients compared to FOXM1^low patients. Following first-line chemotherapy, FOXM1^High patients had a much lower progression-free survival (PFS) compared to FOXM1^Low patients. The presence of extrathoracic metastases, especially bone metastasis, was observed in FOXM1^High patients compared to FOXM1^Low patients [92]. Recently, we have shown that FOXM1 is a critical regulator of SCLC growth and metastasis [93]. It has been shown that C-X-C chemokine receptor type 4 (CXCR4) modulates FOXM1 expression in SCLC, and CXCR4/FOXM1 axis further regulate the expression of EMT regulators including E-cadherin, snail, and Zeb-1. The FOXM1 directly binds to the promoter region of ribonucleotide reductase regulatory subunit M2 (RRM2) and regulate the growth of SCLC cells in vitro and vivo [93]. The outcomes of the study suggested that the microRNA-1 blocks the CXCR4/FOXM1/RRM2 axis to attenuate SCLC growth and metastasis. These findings suggest that the CXCR4/FOXM1/RRM2 axis plays a key regulatory role in SCLC growth and metastasis, and implicated that this axis has a high potential to target overall metastasis of SCLC. Taken together, the findings discussed here suggested that FOXM1 expression has a substantial role in LC (both NSCLC and SCLC) growth and metastasis (Fig. 2) [92].

3.3. FOXM1 in ovarian cancer

Ovarian cancer (OC) is a prominent cancer type affecting women with a 10-year survival rate <40% [94]. FOXM1 upregulation is a key molecular aberration reported in OC. More than 85% of OC patients showed FOXM1 overexpression and activation of FOXM1 or FOXM1-linked oncogenic signaling/molecules, making it one of the most common molecular aberrations reported in OC [95]. Barger et al. demonstrated that FOXM1 develops copy number aberrations in chromosome 12p13.33 at the initial stages of epithelial ovarian cancer (EOC) development and progresses to become high-grade serous carcinoma (the most common type of EOC) [96]. Interestingly, the inactivation of tumor suppressor genes, including TP53 and RB1, has been shown to increase FOXM1 expression [96, 97]. The stage-specific (low to high) FOXM1 and phospho-ERK expression have been reported from low-grade to high-grade OC [98]. High FOXM1 expression was associated with high-grade OC [29]. The differential role of FOXM1 splice variants (FOXM1b and FOXM1c) in various aspects of OC has also been reported; for example, the FOXM1b splice variant was known to enhance invasion and migration, whereas FOXM1c was involved in proliferation, migration, and invasion of OC cells [98]. Further, it has been demonstrated that FOXM1 downregulation via small molecule inhibitors/peptides (thiostrepton or U0126) is p53-dependent in OC cells, suggesting that ERK/FOXM1 mediated oncogenic signaling plays an indispensable role in OC progression and metastasis [98].

Distal-less homeobox 4 (DLX4) is another TF reported to play a crucial role in OC metastasis and is associated with high-grade OC [99]. In addition, DLX5 upregulation has been found to promote the proliferation of OC cells by activating IRS-2-AKT signaling pathways [100]. The relationship between FOXM1 expression and DLX family TFs was not investigated previously in OC. However, Chan et al. have recently revealed that DLX1 is a novel downstream target of FOXM1 and is aberrantly upregulated in high-grade OC [101]. It has been shown that FOXM1-mediated activation of DLX1 enhances the aggressiveness of OC by modulating the TGF-β/SMAD4 signaling pathway [101]. Furthermore, FOXM1 facilitates peritoneal metastasis of OC cells through ZEB1, integrin-β1, integrin-α5, and integrin-αV upregulation and induces adherent to the non-adherent phenotypic transition of OC cells. ZEB1 and integrins are important molecules in cancer cell adhesion and metastasis [102]. Recently, it has been reported that the knockdown of TRIM44 downregulates FOXM1-EZH2 signaling and reduces migration, proliferation, invasion, and colony formation capacity of epithelial OC (EOC) cells in vitro and in vivo OC growth and metastasis [103]. Altogether, FOXM1 is a critical driver in OC, and FOXM1 inhibition underscores the therapeutic implications for the impairment of OC growth and metastasis (Fig. 2).

3.4. FOXM1 in breast cancer

Breast cancer (BC) shows the second highest mortality rate among women, and in 2022 it is estimated that it accounts for ~31% of all the cancers affecting women [104]. Based on molecular phenotype, BC can be divided into four major subtypes, including human epidermal growth factor receptor 2 (HER2+) positive, luminal A, luminal B, and triple-negative breast cancer (TNBC) [105]. Compared to TNBC and advanced-stage BC cases, the patients diagnosed at early stages show good prognoses following surgery and chemotherapy [105, 106]. Treating advanced-stage BC/TNBC patients is difficult, as they have high metastasis and recurrent tumors compared to early-stage BC patients [107]. Tumor heterogeneity and lack of knowledge about molecular targets are the leading cause of difficulty in treating TNBC tumors. Recent findings in BC research identified therapeutic targets with promising therapeutic efficacies for TNBC that improved the disease outcomes, still, overall 10-year survival remains poor [105]. Therefore, novel therapeutic targets must be identified that can be inhibited and thus help improve the disease’s overall outcomes with enhanced OS in various subtypes of BC.

Interestingly, FOXM1 upregulation correlates with poor prognosis in BC [108, 109]. In BC, high expression of FOXM1 was associated with large tumor size, lymph node metastases, lymphovascular invasion, and advanced-stage disease [109]. It was also demonstrated that FOXM1^Low subgroup patients had a better OS than FOXM1^High group [109]. An amazing study on estrogen receptor (ER) positive BC patients suggested that FOXM1^High patients show high recurrence and poor survival [110]. It was also documented that FOXM1 expression is positively correlated with the expression of stemness and EMT markers (CXCR4, E-cadherin, SNAIL, TWIST) of BC cells and decreased tamoxifen sensitivity of BC [110]. Francis et al. provided evidence that FOXM1 is a downstream target of HER2 and identified it as a marker of HER2 upregulation in BC [111]. This provided a rationale for treating HER2+ BC cells with lapatinib (HER2 inhibitor), which reduced the expression of FOXM1 [111]. It has been shown that FOXM1 regulates the expression of estrogen receptor alpha (Erα) in BC cells, and FOXO3a antagonizes this signaling [112, 113]. The transcriptomic sequencing of metastatic TNBC cells revealed a consistent overexpression of FOXM1 in metastatic TNBC cells compared to non-malignant and non-metastatic BC cells [114].

FOXM1 plays a vital role in the EMT of BC cells. It was demonstrated that FOXM1 directly binds to the promoter site of SLUG and activates gene transcription that promotes EMT in BC [115]. Xue et al. showed that FOXM1-mediated SMAD3/SMAD4 activation enhances TGF-β signaling that promotes BC metastases [116]. The TGF-β interaction with SMAD prevented TIF1γ binding to SMAD3 and prevented the ubiquitination-mediated proteasomal degradation of SMAD4 [116]. It was reported that FOXM1 modulates ECM by enhancing the expression of MMP-2, MMP-9, uPA, uPAR, and vascular endothelial growth factor (VEGF) [117]. It has been shown that FOXM1 transcriptionally regulates the expression of VEGF and promotes angiogenesis and cancer cell proliferation in BC [38]. The prominent role of FOXM1 in VEGF regulation suggests its direct involvement in angiogenesis, an essential hallmark of tumor growth and metastasis. Alongside promoting metastasis, FOXM1 has been found to confer resistance in BC cells to paclitaxel, docetaxel, and epirubicin by upregulating XIAP and survivin, which are known antiapoptotic genes [118]. FOXM1 binding to the promoter sequence of platelet-derived growth factor A (PDGF-A) activates the AKT pathway and increases tumorigenesis of BC cells [119]. FOXM1 co-expression with nuclear XIAP and survivin was correlated with poor survival in stage III BC patients compared to patients with low/no expression of XIAP and survivin [118]. A recent study has shown that downregulation of FOXM1 using small molecule inhibitors reduced the tumor growth, metastasis, and EMT-associated markers (snail, slug, vimentin, and MMP2) in TNBC [120]. The outcomes of the study showed that FOXM1 inhibition decreased the tumor growth of TNBC cells in orthotopic mouse models along with metastatic potential [120]. Using a new class of FOXM1 inhibitors named NB-55, NB-73, and NB-115, Ziegler et al. downregulated FOXM1 in ER+ BC and TNBC and reduced the growth of BC in vitro and in vivo with enhanced potency [121]. FOXM1 plays a key role in BC growth, progression, metastases, therapy resistance and overall survival of the patient.

3.5. FOXM1 in Prostate cancer

The second most common cancer in men is prostate cancer (PRC), and worldwide it accounts for the sixth highest number of cancer-related deaths in men [77, 122, 123]. The major driver of PRC pathogenesis and progression is the androgen receptor (AR), and inhibition of AR signaling is used as a primary treatment strategy in the initial stages of PRC [124]. However, in advanced stages, PRC becomes independent of AR-signaling, generally known as castration-resistant prostate cancer (CRPC), and the patients become non-responsive to androgen deprivation therapy [122, 124]. The advanced stage of PRC or CRPC is the most malignant tumor in men associated with metastasis and therapy resistance [77, 125]. Further, frequent skeletal metastasis is one of the major reasons why the 5-year overall survival of PRC patients is reduced to < 3% [126]. Recently, novel targeted therapies have been developed that change the treatment paradigms for metastatic PRC but not all patients show significant treatment benefits [127]. However, the treatment options available for advanced metastatic CPRC are mostly palliative and primarily deal with tumor growth inhibition and pain management [127, 128].

FOXM1 is an exciting target for developing targeted therapies as multiple studies have shown overexpression of FOXM1 in PRC tissues compared to the adjacent normal prostate tissues [129–131]. Pan et al. showed the overexpression of FOXM1 and c-MYC in PRC and established that c-MYC directly binds to the FOXM1 promoter site [129]. The ectopic expression of FOXM1 enhanced the proliferative, migratory, and invasive properties of PRC cell lines, whereas FOXM1 inhibition or downregulation decreased the proliferative and metastatic abilities of PRC cells [129]. A recent report has shown that PRC proliferation depends on FOXM1, and PRC patients with high FOXM1 had a poor prognosis [130]. Mechanistically, FOXM1 regulates EZH2 at the transcript level and plays a critical role in PRC cell proliferation and progression [130]. Jiang et al. showed that long noncoding RNA HOXD-AS1 was highly upregulated in metastatic CRPC patient tumor tissues, cells, and serum-derived exosomes [132]. It has also been shown that HOXD-AS1 containing exosomes were internalized by non-metastatic PRC cells and enhanced the PRC metastasis [132]. Additionally, HOXD-AS1 promotes the distant metastasis of PRC cells in vivo, and it serves as an RNA sponge for miR-361–5p that inhibits the expression of FOXM1 in PRC. Therefore, the outcomes of this study suggest that FOXM1 is a critical component of regulating PRC metastasis and is positively regulated by HOXD-AS1 [132]. Another recent interesting study demonstrated that microRNA-877–5p acts as a tumor suppressor in PRC and directly targets FOXM1 [133]. Ectopic overexpression of miR-877–5p decreased the growth, invasion, and migration of PRC cells suggesting that miR-877–5p nano-formulations could be used as a FOXM1 inhibition strategy to develop future PRC therapies [133]. HIF‑1α enhances the hypoxia-induced EMT via FOXM1 in PRC [134]. The hypoxic environment enhances the expression of HIF-1α and FOXM1, increasing the metastatic potential of PRC cells [134]. Histone H3K4 methyltransferase SETD1A has been involved in the proliferation of metastatic CRPC, and it has been reported that it is overexpressed in metastatic CRPC compared to the primary PRC [135]. It was demonstrated that SETD1A directly binds to the FOXM1 promoter and transactivates FOXM1 in metastatic CRPC [135]. At the FOXM1 promoter region, SETD1A regulates the trimethylation of H3K4 and plays a central role in regulating the expression of octamer-binding transcription factor 4 (OCT4) that contributes to the stemness of PRC cells [135]. High expression of SETD1A-FOXM1 is associated with poor prognosis and plays an indispensable role in the proliferation of metastatic CRPC cells [135].

Docetaxel-based chemotherapy is the first-line therapy for CRPC patients; however, they frequently develop resistance [136]. FOXM1 overexpression has also contributed to docetaxel resistance in CRPC and enhanced autophagy [128]. CRPC cells with high FOXM1 expression exhibited an amplified autophagic flux and a higher number of autophagosomes. The mechanistic analysis further revealed that FOXM1 activates AMPK-mTOR signaling to induce the autophagy pathway that contributes to docetaxel resistance [128]. FOXM1 knockdown (KD) in CRPC cells sensitizes these cells to docetaxel [128]. ATG7 or beclin-1 are essential components of autophagy. Silencing of ATG7 or beclin-1 in combination treatment with chloroquine (which inhibits autophagy) in these FOXM1 overexpressing CRPC cells partially restored the docetaxel sensitivity [128]. Another interesting study showed that FOXM1 directly binds to the Forkhead response elements in the promoter region of kinesin family member 20A (KIF20A) and enhances its expression [137]. High KIF20A expression increased the docetaxel resistance in PRC, and FOXM1-KD induces cell cycle arrest, enhances apoptosis, and decreases the cell migration and invasion of docetaxel-resistant PRC cells. FOXM1 inhibition through thiostrepton sensitizes the docetaxel-resistant PRC cells to docetaxel therapy in vitro and in vivo [137]. The outcomes of the studies discussed here suggesting that FOXM1 plays an essential role in PRC development and progression, and act as a critical driver for docetaxel resistance in CRPC and provides a rationale for the utilization of FOXM1 inhibitors in the combination strategies for docetaxel/drug resistance CRPC patients.

3.6. FOXM1 in Pancreatic cancer

Pancreatic cancer (PC) is estimated to cause the fourth-highest number of deaths in the United States, and it is projected to increase in number further [104]. In the United States, it is projected to take nearly 49,830 lives in 2022 [77, 104]. Pancreatic ductal adenocarcinoma (PDAC) accounts for most PC cases. The lack of a robust detection technique and the late presentation of the clinical symptoms make detecting PDAC at its early stages extremely difficult. Therefore, nearly 80% of PDAC patients are diagnosed at the advanced stage of the disease, and the overall 5-year survival rate of PDAC patients remains less than 11% [138, 139]. Limited therapeutic options, the complexity of the disease, and late detection issues result in poor overall outcomes for PDAC patients [140, 141]. Unfortunately, almost 50% of PDAC patients are diagnosed with distant metastasis with a survival rate of less than a year [141]. Another common outcome in PC patients is the development of cachexia, which further complicates the disease treatment [142]. Driver mutations in KRAS, TP53, CDKN2A, and SAMD4 are the leading factors for PDAC initiation and progression [143–145]. Apart from these frequent driver mutations and targets, other molecules such as FOXM1 have been established as potential therapeutic molecules in various cancer types, including PC. There are ample studies that discuss diverse roles of FOXM1 in PC, but our goal is to discuss the importance of FOXM1 in PC progression and metastasis [32, 43, 146–152].

FOXM1 modulates multiple aspects of PC, including cell cycle progression, cell proliferation and differentiation, DNA-damage repair (DDR) pathways, apoptosis, angiogenesis, tissue homeostasis, metastasis, stem cells, drug resistance, and poor prognosis [32, 39, 147–150, 153, 154]. KRAS mutation is a prominent driver in PC and the Raf/MEK/MAPK pathway, which is downstream of KRAS and has been implicated in stimulating the translocation of FOXM1 to the nucleus to enhance the expression of other target genes [155]. High FOXM1 signaling has been linked to abnormal activation of PI3K/AKT pathway and inactivation of TP53, which coupled with Raf/MEK/MAPK pathway, plays a crucial role in cell-cycle progression and inhibition of apoptosis [155, 156]. Aberrant expression or activation of sonic hedgehog (SHH) signaling has also been observed in various pancreatic diseases such as chronic pancreatitis, PanIN lesions, PDAC tumors, differential diagnosis of PDAC with chronic pancreatitis, and also plays a prognostic role in PC, indicating a crucial part of SHH in the initiation, development, and progression of PC [157, 158].

PC stem cells (PCSCs) are regulated by a wide variety of signaling pathways including PI3K/AKT, Notch, Hedgehog (Hh), NF-κB, JAK/STAT3, PTEN and Wnt/ β-catenin [159]. Overactivation of PI3K-AKT and loss of TP53 function is mechanistically linked with FOXM1 upregulation [156]. Cell surface markers like CXCR4, CD133, CD44, CD24, EpCAM and c-Met all have been used to identify PCSCs but, CD133+ PCSCs have been found to possess much higher tumorigenicity and metastatic potential [159, 160]. FOXM1 is known to interact with the signaling pathways regulating PCSCs which have been mentioned above and induces the expression of the cell surface markers like EpCAM and CD44 [159]. FOXM1 has been observed to maintain the pluripotency of stem cells of embryonal carcinoma by inducing the expression of SOX2, OCT4 and Nanog [161]. This phenomenon can also be true in case of PCSCs, however it needs to be validated through future investigations. The SHH/Gli pathway is also involved in the maintenance of stemness features of PCSCs by activating OCT4, c-MYC, SOX2, Nanog, BMI1, and FOXM1 [40, 162]. Xie et al. showed that FOXM1 transcriptionally activates OCT4 by directly binding to its promoter region in embryonal carcinoma cells [161]. Interestingly, it has been shown that FOXM1 overexpression rescues the expression of TFs associated with pluripotency, including OCT4, SOX2, and Nanog in differentiated cells [161]. The outcomes of the studies discussed here suggest that FOXM1 plays a crucial role in maintaining PC pluripotency. Recently, it has been shown that higher expression of USP28 stabilizes FOXM1, which further transactivates Wnt/β-catenin signaling in PC cells and patient tissue samples [147]. Another exciting study suggested that FOXM1 attenuates the tumor suppressive role of merlin (a protein encoded by NF2 gene) and enhances Wnt/β-catenin signaling [163]. It has been demonstrated that merlin expression reduced the stability of FOXM1 (at the protein level), decreasing the nuclear localization of β-catenin [163]. These studies described the indispensable role of FOXM1 in the modulation of the Wnt/β-catenin pathway and put forward the utilization of FOXM1 as a therapeutic target in the Wnt/β-catenin-mediated cancers [147, 163]. To aide with the stabilization of FOXM1, another important molecule is FAT10, a ubiquitin-like protein that competes with ubiquitin for binding with FOXM1 and prevents the ubiquitin-mediated degradation of FOXM1 [32]. The FAT10-FOXM1 axis shows evidence that it promotes the proliferation of PC cells along with EMT, prevents apoptosis, and imparts gemcitabine chemo-resistance [32]. Bmi1 also plays a significant role in enhancing PC tumorigenicity and increasing the activity of stem cells in the development and progression of PDAC [164]. It has been found that FOXM1 promotes the expression of Bmi1 using c-Myc, thus aggravating PC [164].

On top of that, FOXM1 is known to form a positive feedback loop with HGF/Met, making the PC cells resistant to Met inhibition [165]. It has recently been discovered that autotaxin (ATX) is a transcriptional target of FOXM1, which is essential for FOXM1-mediated proliferation and migration of PDAC cells [166]. ATX was also found to form a positive feedback loop with FOXM1 by downregulating Hippo signaling, enhancing FOXM1 expression and thus facilitating PDAC progression [166]. Dickkopf1 (DKK1) also forms a positive feedback loop with FOXM1, and is involved in PDAC tumor growth [43]. FOXM1 is also known to transactivate uPAR and combined with the FOXM1-caveolin-1 signaling pathway, it promotes EMT and metastasis of PC cells [167, 168]. FOXM1 plays an enormous role in PC initiation, development, and metastasis.

4. Role of FOXM1 in acquiring chemoresistance

FOXM1 is a master regulator for a plethora of genes involved in several cellular processes like the cell cycle, senescence, migration, invasion, oxidative stress, apoptosis, and chemoresistance. Chemoresistance, acquired or inherited, is the foremost hurdle clinicians worldwide face in treating cancer. There are multiple mechanisms by which the cancer cell can become resistant to chemotherapy and recent developments have shown that FOXM1 confers chemoresistance in various cancer types by modulating multiple signaling cascades [110, 169].

Recent advancements have shown that cancer stem cells (CSCs) have enhanced metastasis, invasion, homing, and chemoresistance [40]. Studies have also shown the expansion of the CSCs after acquiring drug resistance, with FOXM1 regulating cancer stemness genes such as OCT4, NANOG, and SOX2 that promote chemoresistance [40, 105, 170, 171]. In prostate cancer, the high expression of cancer stemness markers such as ALDH1, SOX2, and SHH has been reported in prostate post-taxane resistance in vitro [172]. Yuan et al. demonstrated the enhanced expression of cancer stemness markers such as ALDH1, SOX2, and SHH following the taxane resistance in PCa cells, and FOXM1 regulating the expression of these markers through UHRF1 activation [172]. Furthermore, the downregulation of FOXM1 has been shown to reduce the expression of UHRF1 and, eventually, the stemness markers, which sensitizes the taxane resistant PCa cells to taxane [172]. Multiple studies have reported that SOX2 is involved in conferring radio-resistance in several cancer types, such as cervical cancer, head and neck squamous cell carcinoma (HNSCC), and rectal cancer [173–175]. In glioblastoma, FOXM1 confers radio-resistance via SOX2 activation. FOXM1 is involved in the direct transcriptional regulation of SOX2 by binding to its promoter region [176]. The SOX2 gene has five potential FOXO recognition elements in the promoter region. Silencing FOXM1 reduces the expression of SOX2 in both in vitro and in vivo glioblastoma orthotopic models and sensitizes them to irradiation [176].

Analysis of the cancer genome atlas consortium of epithelial ovarian cancer patient cohort has revealed that FOXM1 overexpression may be one of the critical early events in the formation of ovarian cancer. A study by Chiu et al. demonstrated that overexpression of FOXM1 led to increased expression of stemness markers, including BMI1, CD44, NANOG, SOX2, and MYD88 [177]. These ovarian cancer cells also showed higher cisplatin resistance. All these properties conferred by overexpression of FOXM1 result in poor patient prognosis [177, 178]. These reports suggest that FOXM1 confers chemoresistance in cancer cells at the transcription level via regulating the expression of stemness markers (Fig. 2).

The CSC model of drug resistance has been described eloquently in recent years. It suggests that the initial bleak population of the CSCs with drug efflux properties are the only cells that survive after treatment with chemotherapeutic agents [179]. The CSC population in the tumor have higher expression of drug efflux pumps/multidrug efflux pumps such as ATP-binding cassette (ABC) transporters. Multidrug efflux pumps can remove different drugs from cells and confer resistance to many drugs used in cancer therapy or other diseases [180]. Overexpression of ABC transporters is one of the prominent contributing factors to acquiring chemoresistance in cancer [181]. FOXM1 regulates the expression of different ABC transporters in various cancer types. In nasopharyngeal carcinoma, FOXM1 induces paclitaxel resistance by regulating the expression of the ABCC5 gene [182]. FOXM1 binds to the FHK consensus sequence at the promoter region of the ABCC5 gene and activates gene expression [182]. FOXM1 is also involved in the regulation of the ABCA2 transporter. The FOXM1/PHB1/RAF-MEK-ERK axis regulates the expression of ABCA2, which promotes paclitaxel resistance in vivo, resulting in poorer survival in PDAC mouse models [39]. Another ABC transporter regulated by FOXM1 is the ABCG2 transporter. ABCG2 is a prominent stemness marker, and its upregulation has been observed in various drug-resistant cancer types. It has been observed that FOXM1 regulates the expression of ABCG2 in bladder cancer cell lines [66]. FOXM1 directly regulates the activity of ABCC4, a vital drug efflux transporter in retinoblastoma. Overexpression of FOXM1 in the retinoblastoma cells leads to carboplatin resistance [183]. All these studies indicate that FOXM1 acts as a transcription activator of the ABC transporters that causes chemoresistance in various solid tumors. Inhibiting the FOXO3-FOXM1 axis in CCa by treatment with 5-fluorouracil in conjugation with microbial metabolites decreases the activity of drug transporters MDR1, BCRP, MRP2, and MRP7 [184]. A study by Gu et al. revealed that around 69% of relapsed multiple myeloma patients had upregulation of FOXM1 that causes resistance to bortezomib and doxorubicin due to increased drug efflux activity [35]. FOXM1 can also induce drug resistance by activating genes that reverse the effect of drugs. For instance, in breast cancer, FOXM1 alters the microtubule dynamics of the tumor cells by increasing the expression of stathmin protein in vitro [185]. Stathmin negates the effect of paclitaxel, which induces apoptosis by stabilizing the microtubule and causing mitotic arrest. Thus, by activating the transcriptional levels of stathmin, FOXM1 confers paclitaxel resistance in BC cells [185].

The outcomes of these compiled studies establish that FOXM1 employs several methods to induce chemoresistance in solid cancers. The last decade saw a bloom in the field of cancer stemness. Numerous studies have shown that CSCs have the ability of drug efflux that make them resistant to chemotherapeutic treatments. ABC transporters are the key players in multidrug resistance. They have the capacity to efflux a wide range of chemotherapeutic drugs. The afore-mentioned studies indicate that FOXM1 adopts transcription of ABC transporters as one of the main mechanisms to promote chemoresistance in solid tumors. With so many ABC transporters already identified as targets of FOXM1, it is imperative to study the effect of FOXM1 inhibitors in conjugation with chemotherapeutic drugs.

5. FOXM1 inhibitors and targeting status in solid cancers

Currently, few FOXM1 inhibitors are available that are being tested in pre-clinical studies [28]. In recent years, some of interesting structural studies have provided clues for the three-dimensional orientation and architecture of the FOXM1 DNA-binding domain that has accelerated the development of small molecule inhibitors for FOXM1 (Fig. 1) [51, 52]. In addition, some indirect inhibitors of FOXM1 have been identified that targets upstream molecules of FOXM1, such as thiazolidinediones (inhibits SP1), PIN1-FOXM1 blocking peptides, and Diarylheptanoids (inhibits GLI1) are the indirect inhibitors of FOXM1 [28, 186–188]. Derivatives of natural compounds that includes casticin and honokiol have also been found to suppress FOXM1 activity and its downstream effector molecules [189, 190]. Thiostrepton and siomycin have also been found to work as indirect inhibitors of FOXM1 in multiple cancer cell lines, where they prevent auto feedback signal through NFRM loop [191]. Highly specific FOXM1 inhibitors have been developed such as RCM1, FDI-6, 9R-201, NB-55, NB-73 and NB-115 via high throughput screening and phage display methods. [121, 192–194]. These bind to the sulfur of His-287 present in the DBD of FOXM1 [192, 193]. Recently, Katzenellenbogen et al. developed FOXM1 inhibitors that are currently being tested under preclinical evaluations to inhibit FOXM1 in various cancers with high specificity and potency [121, 194]. Despite the multifaceted role of FOXM1 in multiple cancers, not a single FOXM1 inhibitor has reached clinical trial, therefore, there is an urgent need to develop and evaluate FOXM1 inhibitors at a larger scale so that they can be further evaluated in clinical trials. In conclusion, FOXM1 inhibition/inhibitors as a novel targeted therapy have sufficient potential to reach cancer patients.

6. Role of FOXM1 in immunotherapy

The last decade witnessed the importance of immunotherapy as one of the breakthroughs with remarkable improvement in various cancer types [195, 196]. The immune system protects against an array of antigens, however, it may respond differently against self-antigens or show an undesirable immune response. One of the primary reasons behind the limited implications of immunotherapy or immune checkpoint inhibitors (ICIs) is the differential expression of programmed death-ligand (PD-L1) or CD274/B7-H1 [197, 198]. In addition to the expression profile of immune checkpoint molecules, other factors such as tumor microenvironment, T-cell exhaustion, metabolism, and expression of molecules that regulate the transcriptional or translational expression profile of immune checkpoint molecules/targets also modulate the immune response of ICIs in multiple cancers [33, 199–201]. FOXM1 is one of the major TFs that regulate the expression of PD-L1 and modulate the immune response of ICIs (Fig. 2C) [33, 201].

In a recent interesting study, Madhi et al. reported a positive correlation of FOXM1 and PD-L1 expression in LUAD patients [33]. High expression of FOXM1 and PD-L1 was reported in LUAD tissues compared to adjacent normal tissues. It was shown that patients with higher FOXM1 and PD-L1 expression had a lower median OS compared to patients with low FOXM1 or PD-L1 expression. It was demonstrated that FOXM1 is a putative TF that regulates PD-L1 expression and is associated with a poor prognosis in LUAD patients. FOXM1 upregulated the PD-L1 expression by directly binding to the PD-L1 promoter in cancerous cells [33]. It was demonstrated that combination therapy of thiostrepton (TST, a FOXM1 inhibitor) with anti-4–1BB antibody synergistically enhanced immunotherapeutic effects in LUAD, by preventing the overexpression of PD-L1, thus inhibiting immune evasion. TST treatment decreased tumor growth in murine models and increased the number of CD3⁺ T-cells at the tumor site [33]. The outcomes of the study suggested that FOXM1 inhibition increases the sensitivity of tumor cells to immunotherapy [33]. Another interesting study in pediatric refractory solid tumors (neuroblastoma, Ewing sarcoma, rhabdomyosarcoma, and osteosarcoma) along with KOC1 and KIF20A identified FOXM1 as a prominent cancer antigen for anticancer immunotherapy [202]. The study evaluated the safety and the immune response of the NCCV Cocktail-1 vaccine, which is made of cancer peptides including FOXM1, in a phase-I clinical trial. They found that patients with high peptide-specific cytotoxic lymphocyte frequencies for FOXM1, among other peptides, had better progression free survival, thereby providing further evidence that FOXM1 is an ideal target for immunotherapy or to enhance the anticancer immune response of ICIs [202]. A Phase-I study evaluated the use of multiple peptide vaccine in patients with recurrent or persistent cervical cancer [203]. The peptides from FOXM1, Holliday junction-recognition protein, maternal embryonic leucine zipper kinase (MELK), and VEGFR were administered to patients and demonstrated a high T-cell response in the patients administered with FOXM1 and MELK peptides [203].

In addition to the utilization of FOXM1 peptides as vaccine candidates, Kondo et al. identified FOXM1 as an important factor in inducing conventional human CAR-T cells into stem cell memory like T cells (CAR-iT_SCM) [204]. FOXM1 is a downstream target of NOTCH, which is responsible for promoting mitochondrial biogenesis and fatty acid synthesis during the development of iT_SCM [204]. CAR-iT_SCM possesses superior antitumor potential compared to conventional CAR-T cells [204]. It was demonstrated that FOXM1 modulates the metabolic pathways that directs the biogenesis of stem cell memory-like T-cells, suggesting that it has a prominent role in the maintenance and development of immune cells, and thus in mediating the associated immune responses. Therefore, FOXM1 can be utilized to modulate anticancer immune response and the development of CAR-T cells. Shi et al. investigated a novel approach to combat colorectal cancer, using cytokine-induced killer cells (CIKs) based immunotherapy [205]. They showed that CIKs co-cultured with SW480 human colorectal cancer cells downregulated the expression of FOXM1 as compared to SW480 cells alone. This suggest that CIKs inhibit the proliferation and invasion of colorectal cancer cells through FOXM1downregulation [205]. Su et al. developed a novel antigen delivery tool using cytoplasmic transduction peptides (CTP) that transduced FOXM1 tumor antigen into the cytosol of dendritic cells [206]. They discovered that immunity induced by dendritic cells loaded with CTP-FOXM1 significantly inhibited in vivo tumor growth and metastasis of hepatocellular carcinoma [206]. Their results indicate that dendritic cells pulsed with CTP-FOXM1 might be a promising immunotherapy for HCC. Overall, the outcomes of the studies discussed here suggest that FOXM1 has a potential to be translated in terms of immunotherapeutic drugs or FOXM1 inhibitors in combination with ICIs could be used to enhance the efficacies of immunotherapies in various cancer types.

7. Role of FOXM1 in cancer metabolism

Cancer metabolism is an emerging hallmark for various aspects of cancer including growth, progression, metastasis, and chemotherapeutic resistant [207–210]. Otto Warburg summarizes the cancer metabolism as aerobic glycolysis or the Warburg effect, and provided the concept of “injury to respiration”, where he explained that it is a prerequisite to switch a differentiated cell into proliferative cancer cell [211]. Cancer cell utilizes the tricarboxylic acid (TCA) cycle to respire or uses its intermediate as metabolites for growth through a process known as anaplerosis [212, 213]. Multiple metabolic targets have been identified in various cancer types, among them FOXM1 is an emerging metabolic modulator in cancer cells and the cells of tumor microenvironment including T cells [214, 215].

FOXM1 modulates glycolysis and energy production in multiple myeloma [194]. FOXM1 support the growth of myeloma cells through enhancing the glucose uptake, oxygen consumption, and lactate production [194]. Kyuno et al, showed that FOXM1 modulates epithelial to mesenchymal transition (EMT) in a glucose dependent manner and enhanced snail expression in PC [148]. Low glucose conditions increased the expression of claudin-1, and decreased the expression of FOXM1 and snail, whereas high glucose conditions enhanced the expression of FOXM1 and snail in PC cell lines [148]. FOXM1 downregulation induced mitochondrial respiration under high glucose conditions. FOXM1 knockdown or low glucose phenocopies the same activities, which is decreased expression of snail and increased expression of claudin-1 [148]. This shows that FOXM1 regulate cell migratory properties or metastasis through metabolic reprogramming. Another interesting study showed that peptide proteolysistargeting chimeras (PROTAC) of FOXM1 inhibited the migration and invasion of HCC and BC cells, and also decreased the expression of GLUT1 (glucose transporter) and PD-L1, and thus suggested that FOXM1 is involved in cancer metabolism and immunomodulation [216]. FOXM1 directly target the promoter sequences of thymidylate synthase (TYMS, a major metabolic regulator) and regulate its expression in naïve and acquired 5-FU resistant CRC cells [217]. A strong positive correlation between FPXM1 and TYMS has been shown in CRC human CRC tissues specimens [217]. This suggest that FOXM1 mediate 5-FU resistant in CRC through metabolic reprogramming. Recently, Hu et al, showed that mTOR-FOXM1 signaling axis reprograms glucose consumption in the tumor microenvironment of hepatocellular carcinoma (HCC) and modulates the cytotoxic capacities of T-cells and anti-PD1 immune response [215]. It has been shown that the co-treatment of IFNα and anti-PD1 potentiated the immune response. IFNx decreased HIF1α by downregulating FosB transcription, which provided a high-glucose microenvironment as the glucose consumption by cancer cells is decreased, and CD27 (a T-cell costimulatory molecule) through mTOR-FOXM1 axis increases the CD8+ T cell infiltration [215]. The outcome of the study suggests a unique role of FOXM1 in the tumor microenvironment and suggested a new avenue of FOXM1 as a direct modulator of metabolism and immune response.

8. FOXM1 in exosomes and tumor microenvironment

Exosomes are microvesicles with a membrane found mostly in blood, bile and urine and typically have a size range of 40 t0 150 nM [218]. Cancer cell derived exosomes have become a major topic of cancer research and discovered as a way of cell-to-cell communication. Exosomes derived from tumor cells serve as messengers to other tumor and non-tumor cells by delivering proteins, micro RNAs, long non-coding RNAs and DNAs and this is how they modulate local and systemic tumor microenvironment [132]. Exosomal FOXM1 or its mRNA has not been reported yet. However, a FOXM1 related long non-coding RNA known as FRLnc1 has been found to be associated with gastric cancer (GC). GC patients had an enhanced levels of FRLnc1 in their serum exosomes. Knockdown of FRLnc1 was found to reduce proliferation and migration of GC cells (HGC-27) whereas overexpression of FRLnc1 rescued cancer cell growth and metastasis in MKN45 cells. Moreover, high levels of FRLnc1 in GC derived exosomes has been found to correlate with lymph node metastasis and TNM stage [219]. HOXD-AS1 is another exosomal long non-coding RNA (LNC) which is overexpressed in castration resistant prostate cancer and is involved in prostate cancer metastasis in vivo. HOXD-AS1 was mediating metastasis of PRC via modulation of miR-361–5p/FOXM1 axis where it sponges miR-361–5p, a potent tumor suppressor. FOXM1 is regulated by miR-361–5p and by sponging miR-361–5p, HOXD-AS1 enhances the levels of FOXM1, which promotes EMT and consequently metastasis in PRC [132].

Cellular communication via exosomes can remodel tumor microenvironment (TME). One known way to remodel TME is the delivery of LNCs via exosomes, following the cellular uptake exosomes alters gene expression. Jiang et al. have shown that PRC cells release exosomes that transfers HOXD-AS1 and increases the motility of non-metastatic PRC cells. They have also shown in vivo that exosomal transfer of HOXD-AS1 enhanced bone metastasis in PRC in a FOXM1 dependent manner [132]. Modulation of GC cells in the TME using exosomal FRLnc1 was also observed that further increased the proliferation and metastasis of GC cells [219]. Undoubtedly, it can be inferenced that exosomes are the major modulators of TME and cell-to-cell communication; still, modulation of FOXM1 axis through exosomes is in early stages of investigation. However, this is an important area of future investigation that can provide further insights into the metastatic or metabolic related role of FOXM1 in multiple cancer types.

9. Concluding remarks and future perspectives

It has emerged from a multitude of studies that FOXM1 regulates numerous signaling pathways, which have been confirmed in vitro and in vivo studies [220, 221]. Overexpression of FOXM1 has been observed in a plethora of cancer types where it contributes to tumorigenesis, proliferation, and migration (Table 1) [28, 221, 222]. Evidence of poor prognosis linked to FOXM1 has also been shown in many cancer types [28, 223, 224]. Moreover, FOXM1 overexpression is tumor tissue-specific, whereas it is absent in most of adjacent normal tissues [225, 226]. The unique role of FOXM1 in tissue-specific tumorigenesis provides opportunities for designing novel anti-tumor therapy which would be much more specific for cancer cells than conventional chemotherapeutic drugs that mainly target actively dividing cancer cells. There is mounting evidence for the neoplastic roles of FOXM1, compelled researchers to investigate this master regulator in various aspects of cancer regulation.

Table 1:

Cancer associated functional roles of FOXM1 in various cancer types.

Cancer type	Molecule(s)	Mechanism(s)	Outcomes	Reference
Colorectal Cancer	FOXM1	PLAUR overexpression	Promotes metastasis	[75]
		PTTG1 overexpression	Invasion, migration, and metastasis	[78]
Lung Cancer	FOXM1	SNAIL overexpression	Distant metastasis	[87]
		Overexpression of FOXM1	Bone Metastasis	[92]
Ovarian Cancer	FOXM1	FOXM1 Copy number aberration	Epithelial OC development	[96]
		FOXM1 overexpression	Enhancement of OC grade from low to high	[29], [98]
		FOXM1b overexpression	Enhance invasion and migration	[98]
		Aberrant expression of FOXM1c	Proliferation, migration, and invasion	[98]
	DLX1 (Downstream target of FOXM1)	Overexpressed DLX1 modulates TGF-β/SMAD4 signaling pathway	Increases aggressiveness of OC	[101]
	FOXM1	Upregulation of ZEB1, Integrin β1, Integrin-α 5, and Integrin-α V	Induction of non-adherent phenotype in adherent cells	[102]
Breast Cancer	FOXM1	FOXM1 overexpression	Poor prognosis, high tumor size, lymph node metastases, lympho-vascular invasion, and advancement of disease stage	[109]
		FOXM1 Expression	Stemness and EMT marker expression (correlation)	[110]
		Regulation of ERα	Cell proliferation and tamoxifen resistance	[112, 113]
		FOXM1 overexpression	Metastatic progression of TNBC cells	[114]
		Transactivation of SLUG	Promotion Of EMT	[115]
		Activate SMAD3/SMAD4 which enhances TGF-β signaling	Promote metastasis of BC cells	[116]
		Upregulation of MMP-2, MMP-9, uPA, uPAR, and VEGF	Modulation of ECM, tumor growth and metastasis	[117], [228]
		Upregulates expression of antiapoptotic XIAP and Survivin	Confers paclitaxel/docetaxel/epirubicin resistance	[118]
		Transactivates PDGF-A followed by AKT activation	Enhance tumorigenesis	[119]
		Regulate snail, slug, vimentin, and MMP2	tumor growth, metastasis, and EMT	[120]
Prostate Cancer	c-MYC	FOXM1 transactivation	Enhances proliferation, invasion, and migration	[129, 130]
	FOXM1	Regulate transcription of EZH2	Proliferation and Progression of PRC	[130]
	HOXD-AS1	Upregulate FOXM1	Chance the phenotype of PRC cells from non-metastatic to metastatic	[132]
	FOXM1	activates AMPK-mTOR signaling to enhance autophagy	Confers docetaxel resistance	[128]
		UHRF1 activation	Enhances stemness by increasing ALDH1, SOX2, and SHH	[172]
Pancreatic Cancer	FOXM1	Abnormal activation of PI3K/AKT pathway, inactivation of TP53	Cell-cycle progression and inhibition of apoptosis	[155, 156]
	FOXM1	Promotes expression of Bmi-1	Aggravates PC	[164]
	FOXM1	FOXM1 forms a positive feedback loop with HGF/Met	Confers resistance to Met inhibition; makes PC cells more aggressive	[165]
	ATX	Downstream target of FOXM1	Promotes proliferation and migration	[166]
	ATX	Downregulates Hippo signaling and thus upregulate FOXM1	Progression of PDAC	[166]
	DKK1	Forms positive feedback loop with FOXM1	PDAC tumor growth, Possible role in PDAC cell metastasis	[43].
	FOXM1	Transactivates uPAR	Promotes EMT and metastasis of PC cells	[167, 168]

The indispensable and critical functional role of FOXM1 provides a rationale for developing future novel strategies targeting FOXM1 or/and FOXM1-mediated signaling, which might be a viable therapeutic option for various cancer types. Another interesting and potential avenue for targeting FOXM1 is identifying putative or specific ligand-receptor pair(s), which directly modulate FOXM1-signaling, and the major regulators associated with this axis. Identification of FOXM1- regulating receptor(s) will allow researchers to develop alternative targeting strategies by developing small molecules or antibodies to downregulate FOXM1 expression or activity. Another innovative strategy to target FOXM1 or associated signaling would be the development of proteolysis targeting chimeras (PROTACs) against FOXM1 or receptors regulating FOXM1 for proteasomal degradation. The pathways regulating FOXM1 expression and activity remain unexplored or understudied, and this area needs further investigations. High throughput analysis, such as single-cell sequencing, could be a valuable tool to study the FOXM1^High/low cancer cells that may provide a clue or gene signature enriched in FOXM1^High or metastatic cells and can be used to identify the interacting partners or downstream effectors or upstream regulators of FOXM1. As FOXM1 is a major TF that modulates multiple oncogenic transformations, including epigenetic changes, therefore; the comprehensive epigenetic studies such as chromatin accessibility, histone modifications, and promoter modifications in FOXM1^High/Low cells will shed light on major changes associated with or regulating FOXM1 expression.

Recently, it has been shown that FOXM1 can translocate to the mitochondria and impair the oxidative phosphorylation [227]. This impairment of mitochondrial function takes place independent of nuclear transcription, and this has been a novel FOXM1 function [227]. Adequate information is available related to the design and development of new molecules, including crystal structure of FOXM1 DNA binding domain; therefore, this knowledge can be utilized to develop FOXM1 inhibitors targeting DNA binding domain. However, lack of a crystal or co-crystal structure for whole FOXM1 or any other interacting partners further limits the implications of drug development pipelines. Overall, FOXM1 is emerging as a master regulator for most cancers and playing a key role in metastasis and drug resistance. The in-depth knowledge of molecular mechanism(s) or regulators associated with FOXM1 and structural characterization of complete FOXM1 or FOXM1-nucleosome complex will provide the platform to develop therapeutic strategies with translational implications in FOXM1-mediated cancers.

Abbreviations

ASCL1

Acheate-scute homologue 1

CDX

cell line derived xenografts

DDR

DNA damage response

FOXM1

Forkhead box protein M1

GEM

genetically engineered mouse

ICB

immune checkpoint blockade

LC

lung cancer

MYC

Myc proto-oncogene protein

MHC

major histocompatibility complex

NE

neuroendocrine

NeuroD1

neurogenic differentiation factor 1

non-NE

non-neuroendocrine

NFIB

nuclear factor IB

ORR

overall response rate

OS

overall survival

PDX

patient-derived xenografts

PRC

polycomb repressive complex

POU2F3

POU class 2 homeobox 3

PFS

progression free survival

PNECs

pulmonary neuroendocrine cells

SCLC-A

high ASCL1

SCLC-N

high NEUROD1

SCLC-P

high POU2F3

SCLC-Y

high YAP-1

SCLC

Small cell lung cancer

STING

stimulator of interferon genes

TIME

tumor immune microenvironment

TME

tumor microenvironment

TMB

tumor mutational burden

TS

tumor suppressor

VEGF

vascular endothelial growth factors

YAP1

yes associated protein 1