Folate receptor: a potential target in ovarian cancer

Introduction

Folic acid antagonists have been used in the treatment of cancer for more than six decades, since 1948 [1]. Folic acid is a vitamin required for cell metabolism, DNA synthesis, and repair [2]. Two cellular folate uptake pathways have been characterized. The first is mediated by transmembrane-reduced folate carriers, which represent the predominant pathway for folate uptake by normal human cells. Whereas in epithelial cancer cells, membrane proteins called folate receptors are often highly expressed and mediate the folate uptake by tumors. Epithelial ovarian carcinoma (EOC) has the highest expression of these proteins, making the folate receptor an attractive candidate for targeted therapy and targeted drug delivery therapy in several types of cancers, specifically in EOC.

EOC had an age-standardised rates (ASRs) 13.1 per 100,000 inhabitants (2012) [3], with the highest mortality rate of all gynecologic tumors. Prognostic factors in EOC are stage, grade, age, histological subtype, performance status, volume of ascites, extent of residual disease following debulking surgery, and findings at second-look laparotomy [4]. Parameters of the host response, e.g., the presence of tumor-infiltrating lymphocytes [5] or the production of neopterin, a biomarker of immune system activation [6], have also been identified as independent prognostic factors. More research into the prognostic factors is clearly needed as prognostic factors can help to identify patients with poor prognosis and streamline the development of new approaches and drugs.

Currently, the standard treatment for advanced ovarian carcinoma is surgical tumor debulking followed by platinum and paclitaxel-based combination chemotherapy. However, despite the potentially high response rate to the initial chemotherapy ranging approximately between 60% and 80%, the majority of patients will ultimately experience recurrence [7], and only about 44% survive 5 years after the diagnosis. The development of drug resistance, either de novo or induced resistance, significantly limits the long-term efficacy of systemic chemotherapy [8]. For platinum therapy resistant recurrent EOC topotecan, liposomal doxorubicin, etoposide, gemcitabine, vinorelbine, cyclophosphamide, and/or other drugs may be used [9]. Therapeutic approaches that target the host response to neoplasia have so far been limited to the experimental setting [10, 11].

Despite the fact that anti-folic acid chemotherapy has been used for more than six decades, a new area for the exploration of anti-folate therapy has been opened with the advent of targeted therapy and with the development of new targeted approaches in medical oncology. Moreover, folate receptors may be not only the target for anticancer therapy, but also a potential predictive and prognostic biomarker and a target for tumor imaging.

This review summarizes published data concerning the folate receptors, their potential usage as prognostic and predictive biomarker as well as their role in the detection and targeted therapy of EOC, and seeks to define future research directions.

Folate receptor – structure, expression, and function

Folate receptors (FRs) are glycosylphosphatidylinositol-anchored membrane proteins of 38–40 kDa that bind folic acid or folate-drug conjugates with high affinity [12]. FRs are encoded by a multigene family FOLR (namely, FOLR1, FOLR2, and FOLR3), which is localized to chromosome 11q13.3–q14.1 [13]. FRs are membrane proteins that play an important role mainly in the cellular accumulation of folates and antifolates. Folate (pteroyl-l-glutamic acid, vitamin B9) is an essential vitamin that is important for DNA replication and cell division. Three types of folate receptors (FRα, FRβ, and FRγ) have been distinguished [14]. Each of folate receptors has tissue specific distribution and different folate-binding potential [15]. FRα binds folic acid with high affinity and transports folate by receptor-mediated endocytosis [14].

FR has been discovered as a tumor marker of EOC cell line in 1991 [16], and FR expression has been reported in 90% of EOC cases. Subsequently, different levels of FR expression were observed in various human tumors and normal tissues. In contrast to cancer cells, most of the nonmalignant tissues express low to negligible levels of FR, with the exception of the kidney and the lung. FR is limited to the apical membrane of the proximal tubule cells of human kidney and lung pneumocytes [17, 18]. However, in lung tissue, FR seems not to be able to capture folates from the circulation as folate-drug conjugates fail to accumulate in the lungs of patients [19]. Besides EOC, a number of other cancers have the high levels of FR, namely, renal cell carcinomas, lung cancer, and endometrial carcinoma [20]. Moderate expression has been found in the brain, breast, bladder, and pancreatic cancers [20]. In the above-mentioned quantitative radioligand-binding study, more than 50% of pancreatic cancer samples were positive for FR expression, but the study using immunohistochemistry demonstrated only 13% of pancreatic cancer samples as positively staining [21]. This discrepancy could be explained by the antibody specificity, thin slicing of samples, where antigen retrieval is compromised, or by the expression of isoforms other than α-isoforms, as in the immunohistochemical study, anti-FRα antibody has been used. Low FR expression has been reported in the liver, colorectal, prostate, lymphoma, and head and neck carcinomas [20]. Despite the fact that among all other human tissue types ovarian malignancies have the highest FR expression, the level of expression is highly dependent on tumor histology. Almost 100% of nonmucinous primary and metastatic ovarian tumors were found to have high FR expression, whereas mucinous tumors express very low or nondetectable FR levels [20]. Moreover, it seems that the metastatic tumors generally express higher levels of FR compared to the primary tumors [20, 21]. FR expression seems to provide a growth advantage to the tumor by greater folate uptake resulting in rapid cellular growth and division [14]. Consequently, FR protein expression has been found to be associated with tumor progression, dedifferentiation (grade), decreased survival, and platinum resistance [22–24]. Toffoli et al. formulated a hypothesis that FRα might increase the repair of DNA damage caused by platinum via increased folate cellular uptake [23]. Patients with suboptimal debulking surgery had significantly higher tissue FRα expression levels in contrast with patients with optimal surgery. Moreover, a higher expression of FRα was an independent prognostic factor associated with inferior disease-free survival and overall survival of patients [25].

There is evidence that FR might be involved in malignant transformation, tumor progression, and treatment outcomes. In addition to the modulation of treatment outcomes of standard platinum-based chemotherapy, FR is currently studied as a promising potential target for the targeted therapy. Moreover, because FR has been confirmed to be a tumor-associated antigen in ovarian cancer, it may be used for selective targeted delivery of drugs to tumor cells.

Folate receptor as a target for anticancer therapy

The FR represents a potential target for targeted therapy in some human cancers [26]. As the above-mentioned studies have shown a significant association of FR expression in ovarian cancer cells with biological aggressiveness and tumor phenotype, it is clear that FR plays an important role in tumor development and progression. The role of FRα in carcinogenesis may be to regulate folate uptake from serum [27] or generate regulatory signals for tumor growth [28], and therefore, FRα overexpression phenotype is associated with ovarian cancer cell proliferation [29]. The high expression of FRα in ovarian cancer, together with the effect on tumor biology, predisposes FRα for a potential target for the ovarian cancer treatment. Moreover, FRα could suppress the cytotoxic drug-induced apoptosis through the caspase pathway by regulating the apoptosis-related molecule Bcl-2 and Bax [25], and therefore, the combination of FRα target therapy and chemotherapy has been currently intensively studied.

Folate receptor and targeted drug delivery in anticancer therapy

Targeted delivery of anticancer therapy exclusively to cancer cells is certainly an attractive treatment approach in medical oncology. This concept brings several advantages, mainly improvements in efficacy of anticancer therapy and reduction of drug toxicity as high toxicity and insufficient activity are currently the biggest limitations of anticancer therapy [35]. The frequent overexpression of folate receptors in cancer cells may represent a target for tumor-selective drug delivery [36]. In general, two strategies have been shown to be effective for targeted delivery of drugs to FR-positive cancer cells [36]. First is the coupling of an anticancer drug to monoclonal antibody against the receptor. However, in this case, the targeting ligand is highly immunogenic with a range of side effects and with the inability for repeated administrations in the case of development of secondary antibodies. The second approach is coupling of anticancer drugs to folic acid, a high affinity FR ligand. Currently, there are five folic acid-based small-molecule drug conjugates (SMDCs) [26].

Imaging methods

In light of the development of FR-targeted anticancer therapy, the need for quantification of functional FRs has occurred in clinical cancer samples. Several methodical approaches have been reported to analyze the FR-positive cells, namely, polymerase chain reaction detecting mRNA expression levels [48], immunohistochemistry, and radioligand binding assay detecting protein expression levels [20–25]. Quantitative radioligand binding assay has an advantage of sensitively measuring only functional receptors which is very important for FR-targeted therapies. On the other hand, immunohistochemistry can avoid specificity concerns regarding nonmalignant cells as activated macrophages that may contribute to radioligand signaling. Thus, both methods are currently used to established FR expression status in trials.

As several folic acid-based radiopharmaceuticals targeting FR have been developed, some of these agents have been tested in imaging FR via single-photon emission computed tomography (SPECT) or via positron emission tomography (PET) [49].

With regard to PET imaging, the first ¹⁸F-labeled folate has been synthetized for application in 2006 by amide coupling of the prosthetic group 4-[¹⁸F]fluorobenzylamine and native folic acid [50]. Despite the good visualization of FR-positive tumors, the preparation of the radiopharmaceutical was complicated and time consuming and provided a low region-selective product. Therefore, another radiofolate was developed using the copper-catalyzed azide-alkyne cycloaddition (CuAAC; click reaction) resulting in the first ¹⁸F-click-labeled folate in high radiochemical yields, but in vivo animal PET imaging revealed an unfavorable biological distribution profile with a poor signal-to-noise ratio caused by the polarity of the radiofolate [51]. Fischer et al. used the efficacy of CuAAC for ¹⁸F-radiolabeling in combination with the inevitable polarity of [¹⁸F]FDG to enhance pharmacodynamics and significantly enhanced tumor-to-background ratio [52]. Another approach is ¹⁸F-fluorination, which led to the development of 2′-[¹⁸F]fluorofolic acid with promising results in vivo, as a clear-cut visualization of FR-positive KB tumors, and healthy tissues have been shown [53, 54]. Recently, several new ¹⁸F-folate was developed, being available in very high radiochemical yields via a fast and convenient two-step radiosynthesis and showed good in vivo behavior waiting for testing in clinical trials [49, 55].

Radionuclides showing promising results in preclinical in vivo SPECT are ¹¹¹In-diethylenetriamine pentaacetic acid (DTPA)-folates, ⁹⁹mTc-folates, and ⁶⁷Ga-folates [56–58]. The first radionuclide tested in clinical trials was ¹¹¹In-DTPA-folate [59]. However, ⁹⁹mTc-based imaging agents showed more rapid pharmacokinetics and lower cost [19]. In vivo, ⁹⁹mTc-etarfolatide evaluation of FRs correlated to the immunohistochemistry staining results in 61% of 154 patients [60]. The authors suggest that the discrepancy may reflect a different FR expression in the primary tumor and metastatic lesions. The administration of ⁹⁹mTc-etarfolatide was safe as none of the serious adverse events were related to the administration of etarfolatide. ⁹⁹mTc-etarfolatide imaging has been tested in two clinical phase II trials with vintafolide to identify the presence of FR in recurrent/refractory ovarian or endometrial tumors and the correlation of FR expression with response to FR-targeted therapy [9, 61]. In both studies, etarfolatide has been able to identify patients likely to benefit from vintafolide. In addition to histological verification of FRs by immunohistochemistry, SPECT with radiofolates seems to be another approach to verify FR-positive tumors that, in addition, brings the possibility to evaluate the treatment response.

Conclusion

FRα has been shown to be a biomarker of ovarian cancer as it is overexpressed in EOC, but has limited expression in nonmalignant human tissues. The degree of FR expressions corresponds with the stage and grade of the disease. Because of this, FRα seems to be an ideal target for EOC treatment. Moreover, the FR expression levels correlated with the response to platinum-based chemotherapy, and it seems that FR may increase the repair of DNA damage caused by platinum. Together, it is hypothesized that the addition of anti-FR therapy to conventional chemotherapy may be synergistic rather than additive.

Currently, two principal approaches have been studied to use this protein in ovarian cancer treatment. The first is the development of antibodies against FR (farletuzumab) and the second is the use of FR for targeted delivery of anticancer drugs selectively to cancer cells (vintafolide/EC145, EC140, EC131, and EC72). Despite the fact that all these drugs demonstrated antitumor activity in in vitro and in in vivo models as well as a lack of severe adverse events, none of them showed so far a benefit for EOC patients in phase III clinical trials. Based on promising results of phase I and phase II, farletuzumab and vintafolide have been tested in phase III trials recently. However, both of these trials failed, and the questions regarding the efficacy of this approach remain. It is not clear how to design randomized trials to ensure a proper selection of patients that could benefit from anti-FR therapy. Several approaches have been tested, including immunohistochemistry, radioligand-labeled folates in PET and SPECT. However, all these methods are semiquantitative approaches and have limitations with regard to the specificity and sensitivity. From the phase II trial, it seems to be possible that the subgroup(s) of patients who may have benefit from anti-FR therapy exists, but these subgroups still remain to be identified. Moreover, FR has been used for specific drug delivery of anticancer agents as vinca-alkaloids and mitomycin C. However, in clinical practice, the most effective drugs for EOC are still paclitaxel and platinum compounds.

More research is needed to answer the question regarding FR isoform selectivity and the role of FR isoforms in ovarian cancer development and progression. The next point is the role of the patient’s diet as there is a lack of data concerning the interference of anti-FR drugs with the circulating folates. Last, but not the least, is the impact of other potential critical points downstream in FR-signaling pathway that may influence the final effect on the cancer cells. An example is the role of RAS mutations in anti-EGFR therapy in colorectal cancer treatment strategy, where the expression of EGFR is not determinative for the effect of the treatment, but the downstream mutations in the pathway are the most important for the final benefit of the patients in a clinical practice.

In conclusion, the recent shift of farletuzumab and vintafolide into phase III trials had suggested that FR targeting therapy is reaching a final point of the research, and the strategy of identifying high FR expression subgroups of cancer patients would be used in a clinical practice to improve the prognosis of EOC patients. Against the expectations, the phase III trials did not confirm the superiority of anti-FR therapy to chemotherapy alone and even opened up more questions.