RNA Therapeutics to Control Fibrinolysis: Review on Applications in Biology and Medicine

Abstract

Regulation of fibrinolysis, the process that degrades blood clots, is pivotal in maintaining haemostasis. Dysregulation leads to thrombosis or excessive bleeding. Proteins in the fibrinolysis system include fibrinogen, coagulation factor XIII, plasminogen, tissue plasminogen activator (tPA), urokinase (uPA), α2-antiplasmin, thrombin-activatable fibrinolysis inhibitor (TAFI), plasminogen activator inhibitor-1 (PAI-1), α2-macroglobulin, and others. While each of these is a potential therapeutic target for diseases, they each lack effective or long-acting inhibitors. Rapid advances in RNA-based technologies are creating powerful tools to control the expression of proteins. RNA agents can be long-acting and tailored to either decrease or increase production of a specific protein. Advances in nucleic acid delivery, such as by lipid nanoparticles, have enabled the delivery of RNA to the liver, where most proteins of coagulation and fibrinolysis are produced. This review will summarize the classes of RNA that induce 1) inhibition of protein synthesis, including small interfering RNA (siRNA) and antisense oligonucleotides (ASOs); 2) protein expression including messenger RNA (mRNA) and self-amplifying RNA (saRNA); and 3) gene editing for gene knockdown and precise editing. It will review specific examples of RNA therapies targeting proteins in the coagulation and fibrinolysis systems, and comment on the wide range of opportunities for controlling fibrinolysis for biological applications and future therapeutics using state-of-the-art RNA therapies.

1. Introduction

Bleeding or thrombus formation can be exacerbated by an imbalance in haemostasis¹. Congenital bleeding disorders affect more than three million people worldwide, while thrombosis causes >100,000 deaths every year solely in the United States. There is great potential for safely controlling thrombosis and bleeding disorders by modulating fibrinolysis, and RNA therapies can help realize this potential.

Fibrinolysis is initiated upon activation of plasmin, a serine protease, from its zymogen plasminogen². Plasmin cleaves fibrin polymers in blood clots into fibrin degradation products, weakening and eventually clearing the fibrin clot³. Tissue plasminogen activator (tPA) and urokinase plasminogen activator (uPA) cleave the Arg560-Val561 bond in plasminogen to generate plasmin. Fibrinolysis is inhibited by antifibrinolytic factors such as plasminogen activator inhibitor-1 (PAI-1), thrombin activatable fibrinolysis inhibitor (TAFI), α2-antiplasmin, and α2-macroglobulin⁴. Coagulation factor XIII (FXIII) exhibits antifibrinolytic activity by crosslinking α2-antiplasmin to fibrin. A disintegrin and metalloproteinase with a thrombospondin type 1 motif, member 13 (ADAMTS13) contributes to the degradation of clots by cleaving von Willebrand factor (vWF). Fibrinolysis is a highly intricate and multi-level-controlled pathway that normally enables appropriate clot formation and degradation. Thus, modulating fibrinolysis provides many therapeutic opportunities to correct thrombosis and excessive bleeding.

Spontaneous or excessive bleeding occurs in patients with inherited or congenital deficiencies, defects in coagulation and antifibrinolytic proteins, or in platelets⁵. Patients with bleeding disorders have reduced clot stability and elevated clot degradation, increasing bleeding tendency. Inhibiting fibrinolysis is a universal strategy to decrease bleeding in bleeding disorders, trauma, postpartum haemorrhage, surgery, and in other bleeding patients. Three antifibrinolytic drugs have been clinically approved to-date: aprotinin, tranexamic acid (TXA), and ε-aminocaproic Acid (EACA). Aprotinin is a serine-protease inhibitor that has largely been phased out of use, solely used now in select applications such as high-risk surgeries due to an increased risk of mortality⁶. In contrast, TXA and EACA are used extensively for treating bleeding disorders, trauma, menorrhagia, caesarian section, myomectomy, hysterectomy, and other surgical settings. However, both of these small-molecule inhibitors have short half-lives of 2–8 hours, requiring frequent administration⁷.

On the other hand, when clot degradation is impaired thrombosis is exacerbated. Most of the current therapeutics for thrombosis are short-lived and target the coagulation system upstream in the coagulation cascade, such as thrombin and/or coagulation factor Xa. Targeting clot formation increases the risk of severe spontaneous bleeding alongside other life-threatening side effects⁸. While antifibrinolytic proteins are potential targets for anticoagulation, no inhibitors of antifibrinolysis have been approved, in large part because the inhibitors have low specificity or off-target effects in vivo.

RNA therapies are both long-acting and highly specific for the target protein. These features reduce the burden of frequent drug administration associated with treatments and overcome the specificity obstacle that many small molecule- and antibody- based agents face. Various classes of RNA therapies exist, including small interfering RNA (siRNA), antisense oligonucleotides (ASOs), and messenger RNA (mRNA) -based therapeutics⁹. Since the first RNA therapy drug was approved for the treatment of cytomegalovirus retinitis in 1998¹⁰, the FDA has sanctioned nine ASOs¹¹, five siRNA¹², and two mRNA¹³ based therapeutics for clinical use, demonstrating the utility of nucleic acid-based therapies. Advances in drug delivery technology, including lipid nanoparticles and N-acetylgalactosamine (GalNAc) conjugates, have also enabled highly effective targeting of RNA therapeutics to the liver, thus creating a platform for modulating hepatic proteins¹⁴. As most fibrinolytic factors are synthesized in the liver, there is a clear opportunity to use RNA to modulate haemostasis and develop novel therapeutics for a variety of bleeding and clotting disorders¹⁵.

2. Classes of RNA therapies that modulate hepatic proteins

RNA therapies can be tailored to be specific, long-lasting, and safe for enhancing or inhibiting protein production¹⁶. This section will summarize the current approaches used for the studying of biological mechanisms and in the development of therapeutics (Figure 1).

FIGURE 1.

Schematic summarizing hepatic-targeting RNA therapies and their impact on circulating protein levels over time. Curves are a generalized estimate and do not depict real data.

2.1. siRNA and ASOs for gene silencing

State-of-the-art siRNA and ASO can be used to knock down mRNA transcripts of proteins synthesized in the liver and decrease the concentration of circulating proteins in blood. siRNA are double-stranded RNA duplexes usually 19 to 27 base pairs in length designed to target and suppress the translation of the complementary mRNA transcript¹⁷. First described in 2001 following the discovery of RNA interference^18–20, siRNA enact their mechanism by binding to the endonuclease, argonaute-2, where the antisense strand dissociates from the sense strand and forms the active RNA-induced silencing (RISC) complex¹⁷. Binding of the RISC complex to the mRNA of complementary sequence leads to mRNA cleavage or translational repression by sequestration to P-bodies, leading to a reduction in protein synthesis¹⁷. Using siRNA to decrease levels of the mRNA and protein of interest is highly effective and long-lasting, typically requiring smaller doses compared to traditional small molecule inhibitors and less frequent administration. The specificity and mechanism of siRNA also allow proteins with no known inhibitors to be targeted, and the effects of siRNA can be easily reversed with protein replacement products such as FDA-approved plasma-purified proteins. Due to these attractive properties, multiple pre-clinical and clinical trials were initiated in the years following the initial discovery to treat a variety of disorders. This culminated in the FDA approval of the first siRNA drug, patisiran, in 2018, for the treatment of hereditary transthyretin amyloidosis (hATTR)²¹. This neuropathological disease is caused by the production of mutant transthyretin in the liver, which form fibrils that eventually lead to neurodegeneration²¹. Upon administration, patisiran achieves over 80% knockdown of serum transthyretin levels lasting three weeks, resulting in a remarkable enhancement in the quality of life for patients afflicted by hATTR²². Since 2018, four other siRNA therapies have been approved by the FDA: givosiran, for the treatment of acute hepatic porphyria^{12, 23}; lumasiran for the treatment of primary hyperoxaluria type I²⁴; inclisiran for the treatment of atherosclerotic cardiovascular disease²⁵; and vutrisiran also for the treatment of hATTR²⁶. Various other therapies are currently in various stages of clinical development²⁷.

Gene silencing can also be achieved through antisense oligonucleotides (ASOs), which were first employed as a strategy for therapeutic knockdown of hepatic proteins²⁸. ASOs are short (8–50 DNA or RNA nucleotides), single-stranded oligonucleotides designed to bind their complementary target mRNA transcripts²⁸. Binding of ASOs to their target mRNA transcript either represses target protein translation or influences pre-messenger RNA (pre-mRNA) splicing, which is dictated in part by the ASO modification, length, and size²⁹. ASOs designed to lower production of a specific protein bind to the target mRNA, leading to RNAse H1 recruitment, mRNA cleavage, and subsequently transcript degradation²⁹. However, ASOs designed for RNA splicing target sequences often located on or near splice sites. This includes exon-intron junctions, where binding can either enhance or inhibit splice sites recognition, thereby causing intron retention or exon skipping, varied mRNA isoforms, and consequently diverse protein variants²⁹. As ASOs often contain heavily modified backbones, ASOs can be highly stable and be directly administered systemically for hepatic targets, or locally for specific cells or tissues of interest without any delivery system³⁰. The first clinically approved ASO, fomivirsen, debuted in 1998 treating retinitis caused by cytomegalovirus infection in immunocompromised AIDS patients³¹. Since then, seven additional ASOs have been approved for clinical use for a variety of disorders³². Though effective at gene silencing, ASOs also possess several limitations and considerations for their use. Notably, due to the need for repeat or high doses, immunogenicity leading to renal and hepatic toxicity alongside ASO accumulation in specific organs can occur³³. The effects of ASOs can also last up to months, which while may be an advantage in certain cases, can be detrimental in the event of ongoing or potential side-effects if suitable reversal agents are not available. In addition, ASOs take an average of 10 days to deplete the protein of interest, which may not represent a feasible therapeutic approach for different occurrences such as in spontaneous and active haemorrhage. Nevertheless, ASOs represent an important therapeutic strategy to control gene expression, with many currently in clinical development.

2.2. mRNA for protein replacement

State-of-the-art mRNA can be used to express proteins in the liver, either to overexpress a protein, produce mutants of an endogenous protein, or produce exogenous proteins. If the protein can be secreted, the protein will circulate in blood. The potential of RNA therapy extends to overexpressing target proteins, which can be applied as a research tool to simulate specific disease states or as a therapeutic agent to treat deficiency disorders. mRNA therapy enables the in vivo translation of exogenous mRNA to produce a target protein. mRNA exhibits potential for diverse applications, such as the development of vaccines and protein replacement therapies. The most notable application of mRNA therapies is the mRNA-based SARS-CoV-2 vaccines, elasomeran and tozinameran³⁴. Other mRNA-based vaccines have also been explored and are under development, including vaccines for influenza³⁵, Ebola³⁶, Zika³⁷, and HIV³⁸. mRNA-based vaccines are usually administered intramuscularly, where immune cells become transfected with the mRNA and begin producing the encoded antigen, resulting in the formation of antibodies³⁹. With the appropriate delivery system, such as lipid nanoparticles, mRNA can be directed to the liver, turning the liver into a bioreactor synthesizing the target protein. In this manner, physiological to supraphysiological levels of protein can be restored in patients with particular protein-deficiencies. The production of plasma proteins in the liver by current mRNA-LNP technologies typically yields a concentration of approximately 0.1–4 mg/L of protein in blood⁴⁰. To date, no mRNA replacement therapies are approved for clinical use; however, the field of mRNA therapeutics is rapidly evolving and various mRNA-based therapeutic are in the clinical pipeline. One limitation to current mRNA replacement therapies that needs to be overcome is the short half-life of mRNA. Expression of the exogenous protein peaks 6 hours post-injection, and by 48 hours it is undetectable, therefore requiring consistent injections if used prophylactically. Further advances to mRNA technology are expected to improve mRNA stability and lead to longer half-lives on the order of weeks to months⁴¹.

2.3. Self-amplifying RNA for protein replacement

State-of-the-art self-amplifying RNA (saRNA) has the potential to produce proteins in the liver for an extended period of time. saRNA is a synthetic single stranded RNA composed of sequence regions derived from positive-strand viruses, such as alphaviruses, in addition to the coding region for the protein of interest⁴². These viral elements enable the saRNA to self-replicate through the expression of four non-structural proteins that self-assemble to form an RNA-dependent RNA polymerase, which then amplifies the copy number of the encoded target mRNA⁴². This mechanism enables prolonged and sustained expression of the encoded antigen or protein for up to 30 days, thereby allowing saRNA-based therapeutics to be administered at lower doses compared to traditional mRNA-based therapeutics⁴³. Currently, saRNA has been employed for the development of various vaccines, with the first saRNA vaccine for SARS-CoV-2 approved in 2023 in Japan for non-emergency use⁴⁴. Clinical trials are currently ongoing for saRNA-vaccines against influenza⁴⁵. The potential of saRNA in protein replacement fibrinolytic therapeutics is of great interest but limited by concerns of immunogenicity. Following delivery of saRNA to the liver, toxicity, likely caused by viral elements, leads to the clearance of saRNA and inhibition of exogenous protein expression⁴⁵. Current research is working to overcome the immunogenicity of these elements and boost safe protein expression.

2.4. CRISPR for permanent gene knockdown

State-of-the-art CRISPR can knock down proteins in the liver, and thus in blood, permanently. Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) provides a tool to modify genes in a targeted manner, enabling gene functional studies as well as the potential to correct genetic disorders⁴⁶. The CRISPR system is composed of CRISPR-associated proteins (Cas) as well as a guide RNA that is tailored to target a specific DNA region. The guide RNA is a single-stranded RNA molecule about 17 to 20 nucleotides long that directs the Cas proteins, commonly Cas9, to the correct location in the genome. When fully complexed, the CRISPR/Cas9 system binds to the DNA and induces a double-stranded cut⁴⁶. The double-stranded break (DSB) is sensed by the cells and typically repaired via one of two DNA repair mechanisms: non-homologous end joining (NHEJ) and homology-directed repair (HDR). NHEJ involves the direct ligation of the DNA ends, which introduces either insertions or deletions and disrupts the gene⁴⁶. This approach is commonly employed to permanently knock down proteins, thus representing a tool to create permanent gene silencing of fibrinolytic proteins in somatic cells as both therapeutics and novel animal models. In 2023, the FDA approved the first CRISPR therapy, exagamglogene autotemcel (exa-cel) for sickle cell disease⁴⁷. This therapy involves extracting hematopoietic stem progenitor cells from the afflicted patients, followed by transfection of CRISPR/Cas9 components. Transfected cells are then edited, such as through point mutations or indels, via the GATA1 binding region of the B-cell lymphoma/leukemia 11A (BCL11A) gene such that the expression of BCL11A, a transcription factor repressing the expression of fetal hemoglobin, is down regulated^{48, 49}. Reinfusion of the edited cells into the patient thus leads to the production of red blood cells expressing fetal hemoglobin with a reduced sickle phenotype. Patients provided with this therapy became free from the vaso-occlusive crises⁴⁷. Expanding this strategy to other blood disorders could thus correct the genetic mutations involved and restore or modulate coagulation and fibrinolysis.

2.5. CRISPR for precise gene editing

State-of-the-art CRISPR can be used to permanently mutate proteins, including correcting mutations, of proteins produced in the liver, and thus blood. If HDR is performed following a DSB, then template-guided insertions can be made into the genome, allowing for the addition of modified genes⁵⁰. Upon the emergence of CRISPR/Cas9, various other systems have been developed to enhance precision and efficiency. In 2016 base editing emerged, which uses a catalytically impaired Cas9 nuclease fused with a deaminase enzyme. This fusion protein enables the direct conversion of one DNA base pair to another without a double stranded break, thus providing a tool for single-point mutations without insertions and deletions⁵¹. Prime editing uses an engineered Cas9 variant with a reverse transcriptase that directly synthesizes the genetic information into the target DNA, therefore creating a platform for precise insertions. In 2023, the FDA approved lovotibeglogene autotemcel (lovo-cel), the second CRISPR therapy for sickle cell disease⁵². Similar to exa-cel, this therapy is created from autologous blood stem cells. Lovo-cel is composed of a lentivirus tailored to create a variant hemoglobin with a single amino acid substitution that functions similarly to healthy adult hemoglobin^{47, 53}. Various other CRISPR therapies have entered the clinic, including for diseases such as cancer, metabolic and blood disorders, and urinary infections⁵⁴.

3. Approaches to deliver RNA to the liver

Due to the labile and unstable nature of RNA, delivery systems are required to protect them from degradation and successfully transfect target cells. Several transfection agents have been developed and optimized for successful transfection of the liver. These are reviewed below.

3.1. Lipid nanoparticles (LNP)

Lipid nanoparticles (LNP) are the gold standard in non-viral nucleic acid delivery. They are the transfection agent of the three FDA-approved RNA therapies patisiran, tozinameran, and elasomeran^{21, 55, 56}. LNP are typically composed of four lipids: an ionizable lipid; a structural phospholipid; cholesterol; and a pegylated lipid, all of which act together to encapsulate the nucleic acid payload and confer particle stability⁵⁷. Upon intravenous administration, LNP naturally accumulate in the liver by interaction with apolipoprotein E (ApoE), enabling delivery through low-density lipoprotein receptor (LDLR)-mediated endocytosis⁵⁸. The LNP composition, size, and surface by addition of targeting moieties, can be specifically tuned to optimize transfection efficiency and cellular tropism^{57, 59, 60}. For example, LNP incorporating the ionizable lipid, DLin-MC3-DMA, accumulate largely in hepatocytes and Kupffer cells – with over 90% of the injected dose accumulating in the liver four hours post-injection and lesser accumulation in other tissues^{61, 62}. LNP incorporating the ionizable lipid, ALC-0315, accumulate in hepatocytes alongside other hepatic non-parenchymal cells such as stellate cells^60–62. Importantly, LNP enable RNA to be delivered at much lower doses than free RNA to elicit therapeutic benefit, and also extend the duration of effect and decrease immune reactions. The use of newer ionizable lipids have also substantially decreased hepatotoxicity and improved the biodegradability of LNP formulations. This was exemplified by patisiran, where a single administration at 0.3 mg siRNA per kg of body weight (mg/kg) achieves over 90% knockdown of the target protein for three weeks²² with no known reported cases of hepatotoxicity⁶³. Depending on the protein, doses as low as 0.01 mg/kg can also achieve therapeutic knockdown²². In mice, only at higher doses of 5–15 mg/kg will markers of hepatotoxicity, such as elevated alanine transaminase (ALT) and aspartate transaminase (AST) levels, begin to be observed^{64, 65}. The efficacy and hepatic targeting of LNP have been exploited to deliver various other siRNA, mRNA, saRNA and CRISPR therapeutics for the development of vaccines, protein replacement drugs, and gene editing tools.

3.2. N-acetylgalactosamine (GalNAc) conjugates

Another strategy for delivering siRNA consists of chemically linking siRNA to N-acetylgalactosamine (GalNAc) residues¹⁴. GalNAc is a monosaccharide, derived from galactose, used as a ligand to target hepatocytes via galactose-binding asialoglycoprotein (ASGPR) receptors¹⁴. As hepatocytes highly expressed ASGPR, this can be exploited to target proteins and enzymes produced in the liver, including coagulation factors^{14, 66}. Conjugation not only protects the siRNA through steric hindrance and enables efficient cellular uptake, but also increases circulation lifetime and siRNA stability upon administration⁶⁷. FDA approved GalNAc-siRNA include givosiran, lumasiran, inclisiran, and vutrisiran.

3.3. Adeno-associated Viruses (AAVs)

Adeno-associated Viruses (AAV) are small, non-enveloped viruses that belong to the Parvoviridae family⁶⁸. These viruses, typically 20–25 nm in diameter, are composed of a protein capsid which encapsulates a single-stranded DNA molecule that leads to the production of RNA⁶⁸. While AAV are not composed of RNA, they are an important class of gene therapies and an alternative to some RNA therapies. AAV can transfect dividing and non-dividing cells, making them suitable for treating various genetic disorders. In gene therapy applications, the AAV genome is modified to remove the viral genes responsible for pathogenicity, while components essential for packaging and delivering DNA are maintained⁶⁹. The gene of interest can then be inserted in the AAV vector, where transfection into host cells thus leads to particle uncoating, capsid shedding, and release of the DNA genome to produce the protein of interest⁶⁹. AAV can also be used to express short hairpins, a type of RNA molecule that can be designed to target a specific mRNA transcript for degradation, thereby silencing the expression of a specific gene⁷⁰. Upon intravenous administration, AAV mainly deliver their DNA genome to the liver, due to their natural tropism for hepatocytes and highly vascularized nature of the liver, which increases the likelihood of successful gene delivery⁷⁰. In 2017, the first AAV gene therapy, voretigene neparvovec-rzyl, was approved by the FDA to treat biallelic RPE65 mutation-associated retinal dystrophy, a rare retinal disease⁴⁷. To date, eight AAV gene therapies have been approved by the FDA, of which one, valoctocogene roxaparvovec, a one-time gene therapy that uses AAV5 to deliver a functional copy of FVIII to the liver, was approved for the treatment of severe hemophilia A⁷¹.

4. RNA therapies to modulate coagulation and fibrinolysis proteins

4.1. RNA therapies to knock down procoagulant proteins

With the advances in delivering RNA to the liver, there are multiple and clear opportunities to use RNA to regulate different aspects of haemostasis, including using RNA to modulate the proteins and enzymes directly involved in thrombus formation. To this end, several siRNA therapeutics have been developed to silence procoagulant proteins and reduce thrombosis risk. For example, an siRNA-LNP was developed in 2015 to target coagulation factor IX (siFIX). The activated enzyme of the zymogen FIX (FIXa) activates coagulation factor X (FX) to its enzyme (FXa), which in turn activates prothrombin to thrombin, leading to clot formation⁷². Intravenous administration of siFIX allowed up to 99% knockdown of circulating FIX levels, and protected rats from thrombosis⁷². Similarly, siRNA-LNP targeting coagulation factor XII (siFXII) can act as a therapeutic to reduce thrombosis risk without significantly impacting haemostasis^{73, 74}. A GalNAc-siRNA conjugate directed against coagulation factor XI (FXI)⁷⁵ is entering a Phase I clinical trial.

With the development of novel agents silencing procoagulant proteins, the risk of haemorrhage needs to be evaluated, as the inhibition of thrombus formation can potentially lead to an increase in the susceptibility to bleeding. As an alternative strategy to inhibiting procoagulant proteins, silencing proteins involved in clot formation, such as fibrinogen, may be a safer alternative. Indeed, elevated levels of fibrinogen are associated with elevated thrombosis risk⁷⁶; however, there are currently no effective long-term strategies to normalize fibrinogen levels in vivo. Lowering fibrinogen levels using RNA therapy represents a therapeutically relevant strategy to mitigate thrombosis. siRNA-LNP developed against fibrinogen (siFga) lowered circulating fibrinogen concentration down to approximately 10% of normal levels in murine models after a single dose at 1 mg of siRNA per kg of body weight (Figure 2A)⁷⁷. Reduction of fibrinogen with siFga effectively limited clot formation in vitro and prevented the acute phase response induced by complete ligation of the inferior vena cava (IVC), thus resulting in smaller blood clots in vivo (Figure 2B)⁷⁸. Bleeding was not significantly increased in a model of tail-vein or saphenous bleed, suggesting that siFga is an effective tool to tune fibrinogen levels without increasing bleeding risk compared to traditional anticoagulants for managing thrombosis (Figure 2C)⁷⁸. The research and therapeutic potential of siFga has been investigated in various other studies to further elucidate the role of fibrinogen in other diseases^77–81. An ASO has also been developed targeting fibrinogen and used as a tool to further understand the role of fibrinogen in heterotopic ossification and traumatic brain injury^{82, 83}. Overall, these studies highlight the potential of using siRNA-LNP or GalNAc-siRNA conjugates to knock down proteins synthesized in the liver, thus providing an approach which can be applied to fibrinolytic proteins to alternatively manage thrombus formation.

FIGURE 2.

Knockdown of fibrinogen using siRNA and LNP protects mice from occlusive thrombi without increasing bleeding risk. (A) Plasma fibrinogen levels in mice one-week post-administration of siRNA-LNP directed against fibrinogen (siFga), luciferase (siLuc) as a negative control, or no LNPs at all. (B) Thrombus weight in mice treated with either siLuc or siFga following complete ligation of the inferior vena cava to induce thrombus formation. (C) Total bleed time by tail-clip in mice administered siLuc or siFga, or in fibrinogen-deficient (Fga^−/−) mice. ns, not significant; **P<0.01; ****P<0.0001. Data adapted and reproduced with permission in compliance with the respective copyright regulations^{77, 78}.

4.2. RNA therapies to produce procoagulant proteins

mRNA-LNP agents can be applied as a research tool to simulate specific disease states or as a therapeutic agent to treat deficiency disorders. Applied to coagulation, this has been demonstrated whereby administering mRNA-LNP encoding for coagulation factor VIII (FVIII) into a mouse model of hemophilia A increased circulating levels of FVIII by 150% and restored clotting activity⁸⁴. Similarly, administration of mRNA-LNP encoding for coagulation factor IX (FIX) in a mouse model of hemophilia B increased circulating FIX levels and restored normal clotting activity without adverse effects even after repeat administration⁷². These two studies provide the basis for the vast potential of mRNA-LNP as a protein replacement therapy for patients with fibrinolytic protein deficiencies.

4.3. RNA therapies to modulate anticoagulant proteins

In contrast to silencing coagulation to mitigate thrombus formation, siRNA can also be leveraged to knock down proteins with roles in anticoagulation. This can be particularly effective as a therapy to treat bleeding disorders by increasing the probability of clot formation. This strategy has been utilized by a GalNAc-siRNA conjugate targeting antithrombin, called fitusiran, for the treatment of hemophilia A and B^{85, 86}. Antithrombin acts to irreversibly inhibit thrombin by binding to the active site, thereby limiting clot formation and promoting fibrinolysis^86–88. Knockdown of antithrombin can thus facilitate stable clot formation, especially in patients with a bleeding diathesis. Initial Phase I and II trials of fitusiran demonstrated prolonged knockdown of antithrombin up to 10% of circulating levels following monthly subcutaneous injections at a dose of 80 mg⁸⁹. The studies were paused, however, in 2017 and again in 2020 following one fatal and five non-fatal thrombotic complications, respectively. Following the implementation of a revised dosing regimen to maintain antithrombin levels between 15% - 35%, studies have since resumed⁹⁰. In April 2023, two Phase III clinical trials reached their primary endpoint by reducing annualized bleeding in hemophilia A and B patients with and without inhibitors by an average of about 90% compared to patients receiving bypassing agents on-demand⁸⁵.

siRNA targeting protein S has also been explored as a strategy to improve clot formation⁹¹. Protein S, alone or in tandem with protein C, acts as an anticoagulant by inhibiting the activities of the procoagulant enzymes, factors Va (FVa), VIIIa (FVIIIa), and X (FXa)⁹⁸. Similar to fitusiran, subcutaneous administration of a GalNAc-siRNA targeting protein S in mice led to substantial depletion of circulating protein S levels without resulting disseminated intravascular coagulation⁹¹. When administered into FVIII knockout (FVIII^−/−) mice as a model for hemophilia A, mice with reduced levels of protein S observed fewer bleeds following a knee joint injury to model acute hemarthrosis compared to FVIII^−/− mice treated with vehicle alone⁹¹. Collectively these studies demonstrate that siRNA targeting anticoagulation can be a feasible strategy to manage bleeding disorders such as hemophilia A or B. Future studies investigating the knockdown of other anticoagulant factors should be careful to ensure knockdown does not engender thrombosis.

4.4. RNA therapies to modulate fibrinolytic proteins

Beyond targeting coagulation, controlling the degree of fibrinolysis is yet another strategy that can treat bleeding disorders. Unlike knocking down proteins that directly inhibit coagulation, inhibiting fibrinolysis allows clots to form but reduces the rate of clot degradation. Though numerous proteins and enzymes are involved in fibrinolysis, plasminogen is as an attractive target for siRNA-LNP therapy to treat bleeding disorders. Congenital plasminogen deficiency, with plasminogen levels ≤ 150 ng/L, is not a strong risk factor for thrombosis⁹³, and the currently approved antifibrinolytics, TXA and EACA, target plasminogen. Inhibiting plasminogen expression using siRNA is thus expected to be effective and tunable at preventing and treating bleeds, lasting considerably longer than TXA and being highly specific for plasminogen. As proof-of-concept, siRNA-LNP directed against plasminogen (siPlg) knocked down circulating plasminogen levels to < 10% and 25% with a single administration in murine and canine models of hemophilia A, respectively, with sustained knockdown following monthly injections⁶⁵. Importantly, plasminogen knockdown did not result in the pathologies seen in plasminogen deficiency, such as ligneous conjunctivitis or hepatic fibrin deposition, and reduced the annualized bleeding rates with cessation of joint bleeding in the hemophiliac canines⁶⁵. These results therefore suggest that siPlg can be a feasible strategy for effective long-term modulation of fibrinolysis as a treatment for bleeding disorders. An ASO has also been developed for plasminogen (ASO-Plg), but used as a tool to understand the role of plasmin in Alzheimer’s disease progression⁹⁴.

Likewise to knocking down plasminogen, reducing levels of ADAMTS13 is another strategy to potentially manage bleeding. Although not directly involved in fibrinolysis, ADAMTS13 promotes clot dissolution by cleaving von Willebrand factor, decreasing the stability of the clot⁹⁵. Preliminary studies demonstrate that siRNA-LNP directed against ADAMTS13 (siADAMTS13) can induce knockdown of up to 69% in hepatic stellate cells 7 days post-injection (Figure 3)⁶⁰. This is especially notable as there no effective inhibitors for ADAMTS13. Future studies will need to verify the efficacy of ADAMTS13 knockdown in murine models in reducing bleeding, and assess if thrombotic thrombocytopenic purpura (TTP) and adverse symptoms of severe ADAMSTS13 can be avoided. Nonetheless, knocking down fibrinolysis with RNA gene therapy has the potential to decrease the incidence and severity of bleeds.

FIGURE 3.

siRNA-LNP enables knockdown of ADAMTS13 in hepatic stellate cells and reduces plasma ADAMTS13 activity. (A-B) Residual levels of hepatic Adamts13 mRNA (A) along with the corresponding plasma activity of ADAMTS13, measured by fluorescent substrate cleavage (B) from mice one-week post-administration of siRNA against luciferase (siLuc) or ADAMTS13 (siA13) encapsulated in two different LNPs (MC3 or ALC-0315). ns, not significant; *P<0.05. Data adapted and reproduced with permission in compliance with the respective copyright regulations⁶⁰.

4.5. RNA therapies to modulate antifibrinolytic proteins

One final application of delivering RNA to the liver for controlling haemostasis is to modulate antifibrinolytic proteins. This strategy can be particularly advantageous as a method to mitigate thrombosis, as targeting physiological antifibrinolytics can gently promote and enhance clot lysis compared to more aggressively preventing clot formation. An siRNA-LNP targeting coagulation factor XIII (FXIII) has been created⁹⁶. FXIII is a transglutaminase that crosslinks fibrin, thus promoting clot stability. However, FXIII has also been implicated in the development of venous thrombosis by increasing thrombus size and red blood cell retention^{97, 98}. siRNA-LNP directed against FXIII (siFXIII) knocked down circulating levels of FXIII-A by >90% for multiple weeks following a single injection into mice and rabbits (Figure 4A, B)⁹⁶. In addition, when evaluated in a mouse thrombosis ferric chloride model, mice treated with siFXIII had an enhancement in fibrinolysis and a higher relative blood flow following the induction of the thrombus (Figure 4C–E)⁹⁶. Though targeting FXIII appears promising as a tool to increase fibrinolysis, there are many other antifibrinolytic proteins whose activity can potentially also be modulated. To this end, ongoing efforts are underway to develop and fully characterize siRNA-LNP that target proteins such as PAI-1, TAFI, and α2-antiplasmin.

FIGURE 4.

siRNA-LNP for FXIII displays robust knockdown in mice and rabbits and weakens arterial thrombi. (A-B) Knockdown of plasma FXIII-A in mice (A) and rabbits (B) following administration of siRNA directed against FXIII (siFXIII, green) versus siRNA against luciferase (siLuc, purple) (C-D) Representative Doppler graphs (C) and the corresponding quantification of blood flow (D) in mice treated with ferric chloride (black arrow) and Tenecteplase (blue arrow) following administration of siLuc (top Doppler graph) or siFXIII (bottom Doppler graph). *P<0.05; **P<0.01. Data adapted and reproduced with permission in compliance with the respective copyright regulations⁹⁶.

4.6. RNA therapies to modulate platelet function

One exciting application of RNA therapy to thrombosis and haemostasis beyond hepatic targeting is to modulate the function of platelets. Platelets contain procoagulant proteins such as prothrombin, FV, and FXI, as well as antifibrinolytic proteins such as protein S and tissue factor pathway inhibitor (TFPI), which could all potentially be modulated with RNA therapies⁹⁹. Though there are currently no therapies modulating the contribution of platelets to coagulation or fibrinolysis, there is potential for the future. A recent study demonstrated that transfusable platelets can be transfected with mRNA-LNP to express luciferase enzymes (Figure 5A, B)¹⁰⁰. Transfection with mRNA-LNP also did not impact the function and response of the platelets when transfused into a rat model of polytrauma (Figure 5C)¹⁰⁰. Given further optimization, platelets collected for transfusion can be engineered to modulate fibrinolytic proteins, yielding weakened or stabilized clots to treat thrombosis or hemorrhage, respectively.

FIGURE 5.

Platelets transfected with optimized mRNA-LNP can express exogenous NanoLuc and firefly luciferase and can participate in haemostasis in vivo. (A-B) Expression of NanoLuc luciferase (A) and firefly luciferase (B) in platelets four hours following treatment with preoptimized or optimized mRNA-LNP. (C-D) Kidney bleeding time (C) and blood loss (D) in rats inflicted with a kidney puncture wound following transfusion of normal or mRNA-LNP transfected rat platelets. ns, not significant. *P<0.05; **P<0.01. Data adapted and reproduced with permission in compliance with the respective copyright regulations¹⁰⁰.

5. Conclusion

Current clinical standards for modulating the coagulation system, in the case of bleeding and thrombotic disorders, rely heavily on small-molecule and antibody-based inhibitors. These agents have disadvantages including short half-lives and lack of specificity that can cause detrimental side effects and require continuous administration. RNA therapies have potential to address these limitations, as they can be tailored to be specific, long-lasting, and safe for enhancing or inhibiting protein production. These RNA agents can also act as research tools to imitate specific disease states. Recent advances in RNA therapy include the development of siRNA, ASOs, and GalNAc conjugates that target various coagulation and fibrinolytic proteins such as plasminogen, fibrinogen, FXIII and FXI. As more siRNA therapies reach the clinic, other RNA therapeutics are poised to emerge for controlling haemostasis. Protein replacement via mRNA and saRNA agents alongside gene editing via CRISPR holds promise and has far-reaching implications across various fields in medical treatment, precision medicine, and development of novel animal models. We expect that in coming years, novel, permanent-acting therapeutics will emerge with the potential to correct congenital and rare bleeding and thrombotic disorders without the need for continuous treatment.