AKR1C3 inhibitors and exploiters for the treatment and therapeutic monitoring of cancer: a patent review (2020-present).

AKR1C3 has a prevalent role in the progression of many diseases, including CRPC, breast and endometrial cancer, and various leukemias. In the androgen-dependent malignancies, disease progression is attributed to the intratumoral biosynthesis of potent androgens via AKR1C3. Therefore, other androgen-dependent diseases, such as endometriosis, PCOS, benign prostatic hyperplasia and alopecia, may also benefit from the development of AKR1C3 inhibitors and degraders. The development of AKR1C3 inhibitors, and a potent degrader, offers the possibility of employing these compounds in other androgen-dependent diseases and may provide promising alternative approaches for patients who do not respond well to current FDA-approved therapeutics ( Table 1 ). Many of the published patents report highly potent and selective inhibitors of AKR1C3, and with rapidly developing drug resistance a severe issue, particularly in the treatment of metastatic cancers, the potential use of these inhibitors to not only provide a new target for single agent intervention, but also to induce re-sensitization to the standard of care offers an enticing possibility to extend not only patient survival, but by reducing dosing of the chemotherapeutic agent to reduce adverse effects and subsequently improving quality of life for the patient. Of particular promise is the report that degradation of AKR1C3 brings about degradation of AR-v7, a common AR mutant, itself responsible for chemotherapy resistance and long deemed ‘undruggable’. This introduces the tantalizing potential of one compound targeting two major causes of ARSI resistance and providing an alternative, and potentially robust, approach to overcoming drug resistance to these agents. The patent highlights the conceivable potential for further degraders to be investigated for the targeting of AKR1C3 for androgen-based cancers, such as prostate cancer (PC), CRPC, endometrial and breast cancers, in which AKR1C3 overexpression is characterized. Although highly promising, PROTAC-based degraders often suffer from poor pharmacokinetic properties with the compound exceeding the Ro5, therefore future PK profiling will give greater insight into PROTAC applicability into clinical translation. Encouragingly the first PROTAC based compound gained clinical approval in May 2026 in the U.S. The potential for further advancements in this field is significant, considering the rapid development of many compounds that not only inhibit AKR1C3, but also utilize its activity to activate prodrugs, allowing for the generation of novel compounds that target proteins outside of the androgen-dependent cascade. This has been shown prevalently in NSCLC, wherein AKR1C3 action is exploited to cleave the aziridine moiety of a prodrug compound and induce DNA cross-linking. Whilst opportunistic, these therapeutic strategies are not without concern, with off-target effects observed with DNA cross-linking, therefore it is imperative that selective targeting of cancer cells is achieved to minimize toxicity and ensure that the therapeutic window remains as wide as possible. Overall, these potential applications further highlight the importance of AKR1C3 and encourage the exploration of AKR1C3-activated compounds in novel fields that are not necessarily androgen-dependent. The observation of correlation between AKR1C3 and PSA levels provides a promising approach not only for monitoring PC progression but also provides an insight into patient survival and potential response toADT. This would greatly improve the diagnoses rate of PC in the clinic and would allow for more precise, and more reliable methods for PC screening, which in turn will allow for more accurate diagnoses and better dosing regimens. The prospect of an AKR1C3 inhibitor becoming a theragnostic, capable of both inhibiting AKR1C3 to exert therapeutic effect but also tracking expression to offer real-time monitoring of response is tantalizing. Overall, the outlook for targeting AKR1C3 is promising with many inhibitors and the first degrader reported, alongside detection and IHC methods developed as biomarkers for disease monitoring, yet the future challenges of clinical translation will become more apparent in the near future. This will allow for expansion of drug discovery into areas which have not been considered and, in turn, will open more opportunities to tackle drug resistance and ultimately benefit patients.

Since 2020, four patents have been published, claiming various compounds inhibiting AKR1C3 (WO2022/234193 [ 44 ], US2020/0085791 [ 45 ], WO2024/030627 [ 46 ] and WO2024/035780 [ 47 ]). First published in 2017 as WO2018/114677A2, Bayer republished US2020/0085791, where the inventors reported the discovery of their [phenylsulfonyl)octahydro-epiminoisoindol-yl](1H-1,2,3-triazol-5-yl)methanone series ( Figure 2 )[ 45 ]. The inventors report the discovery of 69 compounds capable of inhibiting AKR1C3. Recombinant AKR1C3 protein assays were performed, resulting in the identification of lead compounds 1, 2, 3 , and 4 , all possessing IC 50 values of 2 nM, with all compounds of the chemotype expressing an IC 50 <1100 nM. In addition to this patent, Bayer was granted international patents (EP3464288B1, EA036824B1 and CA3025420C) for their initial patent application WO2017/202817A1, where the inventors reported the discovery of their [8-(phenylsulfonyl)-3,8-diazabicyclo[3.2.1]oct-3-yl](1H-1,2,3-triazol-4-yl)methanone series [ 48 ]. Within this patent, the inventors report the development of 76 compounds and their respective IC 50 values towards AKR1C3. Of these compounds, the most potent compounds were found to be 5 and 6, which report IC 50 values of 0.5 nM ( Figure 3 ). Bayer also was granted US2020/0079758A1, originally filed as WO2018/114670A1, which reports the discovery of their [4-(phenylsulfonyl)piperazin-1-yl](1H-1,2,3-triazol-4-yl)methanone series [ 49 ]. For this series, the inventors report the development of 140 compounds and their respective AKR1C3 IC 50 values, with 17 of the compounds reporting an IC 50 value of 1 nM. Patent WO2022/234193, filed by Forendo Pharma LTD, reports the development of 173 compounds that inhibit AKR1C3 [ 44 ]. These compounds commonly exhibit a piperidine moiety, as part of a conjugated urea, as well as an alkenyl-cyano functional group, which has been shown to be key to the compound’s inhibitory activity ( Figure 4 ). All compounds were evaluated for their inhibitory activity against AKR1C3 recombinant protein and selectivity determined over AKR1C2 and 17β-HSD2. The related protein AKR1C2 plays a key role in the inactivation of 5α-DHT, a potent androgen known to facilitate aggressive CRPC proliferation, and so inhibition of AKR1C2 over AKR1C3 can induce further cell proliferation as previously alluded to 50]. Additionally, 17β-HSD2 can convert potent androgens, such as estradiol, testosterone and 5α-DHT into less potent forms; estrone, androstenedione and 5α-dihydrotestosterone respectively [ 51 ] [ 52 ]. Therefore, it is imperative that novel AKR1C3 inhibitors can selectively inhibit AKR1C3 over AKR1C2 and 17β-HSD2. Each of the 173 compounds were investigated for their inhibitory activity towards these three proteins. All of the compounds possessed greater than 78% inhibition of AKR1C3 at 500 nM concentration after 2 hours of treatment, with most compounds reporting >95% inhibition. The maximal % inhibition reported was 103% ( 7, 8 and 9 ), with the >100% inhibition attributed to errors associated with performing fluorescent studies with recombinant proteins. For most of the compounds, inhibition of AKR1C2 and 17β-HSD2 was low, with % inhibitions reported as low as 0-2% for AKR1C2 and 2% for 17β-HSD2. For the least selective compounds ( 11 and 12 ), inhibition for AKR1C3 over AKR1C2 and 17β-HSD2 is still significant, with 12 at 500 nM concentration possessing an AKR1C3 inhibition of 99%, as compared to 29% for AKR1C2 and 50% for 17β-HSD2, and 11 at 500 nM concentration possessing an AKR1C3 inhibitory activity of 99% compared to 42% and 24% respectively. On the other hand, the most selective compound, 10, at 500 nM concentration possessed an AKR1C3 inhibitory activity of 96%, compared to 5% for AKR1C2 and 0% for 17β-HSD2, highlighting the ability for their compounds to inhibit AKR1C3 selectively. Patent WO2024/030627 reports the development of a library of AKR1C3 inhibitors ( Figure 5 ) [ 46 ]. For this structure-activity relationship (SAR) study, the inventors investigated the effects on activity and selectivity over AKR1C1, AKR1C2 and AKR1C4 following changes to the western ring substitution patterns. First, they investigated the activity changes upon modifying R1 by introducing various aromatic substituents as compared to their reference compound, where R1 was a terminal prenyl chain. This reference compound reported an IC 50 of 70 nM and a >2850-fold selectivity towards AKR1C3 over AKR1C2. Replacement of the initial prenyl chain with an unsubstituted phenyl ring ( A1-a ) afforded greater inhibition activity of 40 nM, but with a slight decrease in selectivity over AKR1C2 (>2500-fold compared to >2850-fold). Introduction of para -fluoro, para -chloro and para -bromo substituents to this new phenyl ring saw slightly decreased activity and a range of selectivity values over AKR1C2, reporting selectivity of 600-fold, 333-fold and 277-fold respectively. However, selectivity over AKR1C2 is improved when the fluoro-substituent is moved to the meta position (960-fold) and more when at the ortho position (>1430-fold). Alternatively, addition of electron-donating substituents at the para position of the new phenyl ring (methyl, ethyl isopropyl, methoxy, cyano and phenoxy) provided reduced activities and decreased selectivity as compared to the unsubstituted phenyl ring. Computational modelling indicated the presence of an open pocket adjacent to the steroid binding site, and hence, the inventors investigated the effect of fused bicyclic ring structures that would occupy this pocket upon binding. This allowed the identification of A1-r , consisting of a triazolopyridine moiety, with an improved AKR1C3 IC 50 value of 51 nM with an acceptable 1216-fold selectivity over AKR1C2. Extending the length of the bicyclic system with a biphenyl moiety saw comparable IC 50 value of 50 nM, but a significant decrease in selectivity (580-fold) and any attempts to introduce bulkier aromatic moieties was also met with significantly decreased IC 50 values and selectivity. Next, the inventors investigated the SAR of R2, by replacement of the carboxylic acid. Replacement with a terminal amide ablated the activity (IC 50 28 μM) and while the introduction of the terminal cyano moiety saw retention of AKR1C3 activity (with a bromo-substitution at R1), upon replacing R1 back to the prenyl moiety, activity was attenuated (170 nM to 1.4 μM), identifying that the carboxylic acid moiety is vital for AKR1C3 activity. This led to the identification of the lead compounds A1-r and A1-a . To further understand the binding modalities of these inhibitors, additional computational modeling was conducted on lead compounds A1-r and A1-a , which predicted that key hydrogen bonding interactions occur between the central amide carbonyl and the TYR55 and HIS117 amino acid residues of AKR1C3 ( Figure 6 ) [ 53 ] In addition, with the carbonyl directly attached to the central phenyl ring, the hydrogen bonds are likely stronger and report a much shorter hydrogen bond distance, which attributes to its superior activity and selectivity. The carboxylic acid was predicted to bind in the sub-pocket 3 region and form strong hydrogen bonds with TYR24 and a salt bridge to ARG226, which highlights the requirement for the carboxylic acid, explaining the loss of activity upon replacement at R2. Finally, the [ 1 , 2 , 4 ]triazolo[1,5- a ]pyridine-6-yl moiety at R1 was predicted to form strong hydrophobic interactions with residues PHE311, TYR317, PRO318 and TRY319, which may contribute to the increased activity and selectivity over AKR1C2. In addition, the inventors also investigated the in vitro plasma and microsome stability envisioning the possibility of a prodrug strategy utilizing the methyl ester of the carboxylic acid. Gastrointestinal fluid stability studies revealed that both the methyl ester and the free acid were stable for 1 hour in simulated gastric fluid, and 2 hours in simulated intestinal fluid, reporting that oral administration may be favorable for further studies. Additionally, the methyl ester was reported unstable in mouse plasma and within 30 minutes, complete conversion to the active free acid was observed, making the prodrug strategy a promising approach to dosing. This was further supported with stability in mouse and human liver microsomes, with the free acid showing no CYP degradation, whereas the prodrug saw 25% degradation, in which the free acid was identified as the major metabolite. Based on the favorable activity, selectivity and in vitro stability data, the prodrug was advanced into an in vivo pharmacokinetic (PK) study, which reports that exposure of the free acid metabolite was significantly higher when dosed as the prodrug, than when dosed directly as the free acid (AUC o-t and AUC o-INF 17-fold and 13-fold) as well as the C max reporting a 66-fold increase for the prodrug metabolite. Finally, the antitumor activity was investigated. Upon treating NSG mice implanted with 22Rv1 cells with either 25 mg/kg or 50 mg/kg once a day intraperitoneally, a clear dose-dependent relationship was observed, with 30% and 45% tumor volume reduction respectively. These studies highlight a promising candidate for the treatment of CRPC, by targeted inhibition of AKR1C3 with high potency and selectivity, along with favorable PK properties and in vivo antitumor activity. Related to WO2024/030627, the same inventors also filed patent WO2024/035780 [ 47 ], in which they report the development of a novel AKR1C3 degrader, capable of dual degrading AKR1C3 and AR-v7, a common splice variant of the AR [ 27 ]. This variant has been shown to directly complex with AKR1C3, leading to ARSI resistance and allowing for tumor growth despite low levels of circulating androgens in PC [ 54 ] [ 55 ] [ 56 ]. This discovery was attributed to the observed degradation of AR-v7 upon treatment with compound A1-a , originally discovered as part of the SAR for AKR1C3 inhibitors in patent WO2024/030627, where 1 μM treatment for 72 hours saw 50% degradation of both AKR1C3 and AR-v7 [ 46 ]. Hence, the inventors sought to exploit this phenomenon, leading to the generation of degrader 13 ( Figure 7 ). Proteolysis targeting chimeras (PROTACs) are heterobifunctional molecules consisting of a warhead component (for binding to the protein of interest), an E3 ligase ligand component (for recruitment of the desired E3 ligase), and a linker component for binding the two ligand components together. In doing so, the warhead can bind to the target protein, bringing the protein into close proximity to the E3 ligase, allowing for subsequent ubiquitination and degradation via the 26S proteasome mechanism [ 57 ]. These degraders have many advantages over small molecules, including catalytic activity allowing for smaller dosages as well as the ability to target previously identified “undruggable” proteins, such as AR-v7 [ 58 ] [ 59 ]. However, PROTAC degraders are also plagued with issues, mainly in PK properties as they often exceed Lipinski’s Rule of 5 (Ro5) and are often subject to rapid metabolism and hence, higher clearance [ 60 ] [ 61 ]. The reported degrader consists of the AKR1C3 warhead, as previously identified in WO2024/030627 [ 46 ], bound to a triazole-PEG2 based linker, attached to Lenalidomide, a known substrate for the Cereblon E3 ligase. Docking studies predicted binding of the warhead component to the SP1 pocket of AKR1C3 would be achieved and showed sufficient length of the PEG2 linker, which allows for sufficient solvent exposure of the Lenalidomide for recruitment of the Cereblon ligase. When evaluated in vitro , the degrader is equipotent to compound A1-a , with a reported IC 50 of 49 nM in 22Rv1 CRPC cells, however, degrader activity was decreased at higher concentrations (>1 μM). This was attributed to the “hook effect”, a common phenomenon observed with PROTAC degraders in which the molecules are unable to form the required ternary complex for inducing degradation. Instead, unfavorable binary complexes are formed and no degradation is observed [ 62 ]. Additional time studies were performed and showed time-dependent degradation of both AKR1C3 and AR-v7, with maximal degradation observed at 72 hours at 10 nM concentrations, with notable degradation occurring at 4 hours post treatment. Further degradation studies report a DC 50 (concentration required for 50% protein degradation) of 52 nM for AKR1C3, making this compound the most potent AKR1C3 degrader reported to date. When compared to other AKR members, degradation towards AKR1C1/C2 as well as AR-v7 was also observed, reporting DC 50 values of 49 nM and 70 nM respectively. Confirmation of PROTAC functionality was also confirmed with mechanism of action studies, including pretreatments with MG132, a known proteasome inhibitor [ 63 ]. Upon pretreatment with MG132 at 3 μM, PROTAC activity was completely attenuated, confirming proteasome involvement. Mechanistically, this degrader is promising, yet future directions may look to diversifying E3 ligase ligands present on the molecule, ensuring that the optimal ligand is incorporated for maximal AKR1C3 degradation. While a promising advance, no data was reported on the degrader’s PK profile, a known issue for PROTACS, which often suffer from poor oral bioavailability [ 64 ]. Upon procurement of PK data, insight into in vivo proof-of-concept will be required before translational potential can be assessed. With 4 patents having been filed in the US between 2000 and the time of writing directly targeting AKR1C3, through inhibition or degradation, this shows a promising outlook for future SAR studies. With AKR1C3 overexpression prevalent in many cancers as well as PCOS and endometriosis, targeting AKR1C3 is a potentially viable approach for therapeutic development, particularly for those patients with noted chemoresistance to currently approved chemotherapeutics. While many patents have been filed for direct targeting of AKR1C3, other patents have also been filed that utilize AKR1C3 activity to reduce compounds into their active metabolites (WO2021/005586 [ 65 ], WO2024/078392 [ 66 ], WO2024/023666 [ 67 ], WO2024/023641 [ 68 ], US2024/0316079A1 [ 69 ] and WO2025/214434 [ 70 ]). While more patents have been filed for AKR1C3-activated compounds than direct inhibitors, many of these patents are derived from similar projects including WO2021/005586, WO2024/023666 and WO2024/023641 from Novartis; and WO2024/078392 and WO2025/214434. In 2021, Novartis filed a patent application for the discovery of various tricyclic AKR1C3-dependant lysyl-tRNA synthetase (KARS) inhibitors (WO2021/005586) [ 65 ]. InNF-E2 p45-related factor 2/Kelch-like ECH-associated protein 1 (NRF2/KEAP1) mutated cancers , AKR1C3 has been shown to be upregulated and hence, has been used previously as a biomarker for disease progression [ 71 ]. Due to this upregulation, the inventors proposed that the protein can be utilized to activate their compounds, through reduction of a ketone into the active alcohol. Within this patent, the inventors utilize AC 50 (half maximal activity concentration) to allow for potency values to remain consistent throughout the patent. In their SAR, they report the synthesis of 172 compounds and screened their ability to induce cell death in H460 cells, chosen due to the high expression of NRF2 and AKR1C3, compared to an AKR1C3 null cell line HARA. For this assay, AC 50 values represent the half maximal inhibitory concentration. The AC 50 data reported is extremely promising with many compounds reporting potency of <1 μM, with some compounds achieving values as low as 100 pM. Upon treatment with the unreduced ketones, compounds presented greater activity in the AKR1C3 cell line and minimal activity within the AKR1C3 null cell line, whereas compared to the reduced alcohol derivatives, potency was often retained in both cell lines, confirming AKR1C3-mediated conversion to the active metabolite. From this, the inventors identified compound 14 , the parent compound of 15 , as the lead compound with a reported AC 50 of 83 nM in H460 cells ( Figure 8 ). Compound 15 saw an increased potency of 21 nM, confirming that a prodrug strategy can be pursued. To confirm that AKR1C3-mediated reduction of compound 14 was achieved, various cell lines were treated with the compound and their AC 50 values towards inducing cell death were determined. This showed that as AKR1C3 transcripts were increased, potency of 14 also increased, with an AC 50 of 52 nM reported in NCI-H1944 cells, which reports >3100 transcripts per million of AKR1C3 by RNAseq. Next, the inventors investigated the ability of the compounds to inhibit KARS, as a result of the AKR1C3 reduction. One of the key roles of KARS is the catalytic conversion of ATP into AMP, therefore, to measure KARS inhibition, the inventors performed a fluorescence polarization immunoassay on the detection of AMP. The AC 50 values reported represent increased polarization within the assay, due to the absence of AMP, as a direct result of KARS inhibition. This assay found that the parent ketone 14 was inactive against KARS, however the reduced alcohol product 15 significantly increased potency to 9.1 nM, confirming that the AKR1C3-mediated prodrug strategy can be exploited to target KARS. With compound 14 in hand, anticancer efficacy was investigated in an NCI-H1944 xenograft mouse model. They found that when dosed at 75 mg/kg once daily, via oral administration, tumor stasis was achieved, and when the dose was increased to 150 mg/kg, a 98.5% regression was observed. When performed in an NCI-H460 xenograft model (lower AKR1C3 expression levels), a higher dose of 300 mg/kg was required to achieve tumor stasis, confirming the key role AKR1C3 has in the reduction of the compounds, and hence, generating the more active metabolite. A second patent was filed, for the crystalline forms of compound 14 (WO2024/023666) [ 67 ]. In this patent, the inventors investigated the stability of their compound in various conditions, including changes in humidity, temperature, solutions and pH values. Crystallized 14 possesses high stability at high humidity (>75%), temperature (80 °C) and water stability, which allows the manufacturing and storage of 14 in a safer and more reliable manner, regardless of atmospheric conditions. With regards to pH values, only 5% degradation was observed at pH 1 after 1 week, whereas in basic conditions, degradation was reported to be <1%, even when stored at 80 °C. Changes in humidity, temperature and light were also observed with maximal degradation reported at 5% in acidic conditions at 80 °C, allowing for the compound to be stored for extended periods of time, in a range of conditions. Novartis also filed WO2024/023641, investigating the dosing regimen for a potential upcoming trial of compound 14 , with their primary objective to investigate the safety and tolerability of the drug in Non-Small Cell Lung Cancer (NSCLC) and to determine maximum tolerated dose for future studies [ 68 ]. As a result, these patents provided considerable insight into the role of AKR1C3 in the metabolism of potential drug compounds and how AKR1C3 overexpression may be exploited to achieve prodrug cleavage. In 2016, Threshold Pharmaceuticals published WO2016/145092, which reported the discovery of various DNA alkylating agents for the treatment of various cancers, utilizing AKR1C3 action to cleave the prodrug moiety [ 72 ]. Within this patent, they reported the discovery of AST-3424 (originally filed as TH-3424 ), a novel prodrug in which AKR1C3 reduces the nitro-group into the respective oxime, and subsequent cleavage to the cytotoxic aziridine ( 2660 ), which induces alkylation and cross-linking of DNA ( Figure 9 ). Since this discovery, the licenses for AST-3424 were sold to OBI pharma and have been progressed into phase I trials in the USA ( NCT04315324 and NCT03592264 ). The former trial is ongoing; however, the latter trial has since been terminated due to minimal evidence of clinical activity in the tumor types enrolled in the study. Since then, OBI pharma filed for patent application US2024/0316079, claiming activity of the compound in various in vitro and in vivo models, as well as confirmation of activation by AKR1C3 [ 69 ]. For these models, the tumor growth inhibition (TGI) was reported as the difference between the changes in tumor sizes of the treated group, as compared to the control group. Hence, values exceeding 100% inhibition show tumor shrinkage, as compared to negative values, which show tumor growth. In a HepG2 orthotopic xenograft model, dose-dependent TGI was observed, with a maximal inhibition of 101.2% observed with only 5 mg/kg once weekly, for two weeks. This trend was also observed for the H460 subcutaneous xenograft model, with 88% inhibition observed with a 2.5 mg/kg dose once weekly, for two weeks. The compound was also investigated for its ability to be used in combination with known chemotherapeutics in a range of cancer xenograft models, including VCaP (prostate cancer), SNU-16 (gastric cancer) and A498 (renal cell carcinoma). The VCaP model experienced an increase in TGI from 148%, when dosed alone, to 158% when dosed in combination with prednisolone. For the other models, a similar trend was observed from 73% to 88% in A498 when dosed in combination with sunitinib. The results not only show the potential for AST-3424 to be utilized as a monotherapy, but also the potential for its use in combination with other known chemotherapeutics. Finally, the inventors investigated various patient-derived xenograft models, including those for pancreatic cancer (PA128), gastric cancer (GA6201), high AKR1C3 expression lung cancer (LU2505) and low AKR1C3 expression lung cancer (LU2057). The TGI was reported as 77.4% for PA128, 110% for GA6201 and 105.2% for LU2505, while the LU2057 model reported a tumor growth of 20.9%. The prevalent difference in tumor growth inhibition is indicative of the AKR1C3-activating mechanism as suggested and provides a promising insight into alternative approaches to DNA alkylating agents. Building off the success of AST-3424 , other phosphoramidate compounds have been claimed as alternative DNA alkylating agents via activation by AKR1C3. Anrui Biomedical Technology filed WO2024/078392, reporting the development of 569 compounds, which upon reduction by AKR1C3, released the active phosphoramidite, inducing alkylation and subsequent DNA cross-linkage [ 66 ]. These compounds expand the SAR, by replacing the central aromatic ring of AST-3424 featuring the reducing nitro group, with various bicyclic ring structures which engender greater fit into the hydrophobic SP1 pocket of AKR1C3. The inventors performed IC 50 screenings, where the IC 50 represents H460 cell growth inhibition, and categorized the compounds in various groups, with many reporting values <10 nM. The compound activities were further shown to be reduced upon pretreatment with ASP9521, a known AKR1C3 inhibitor, confirming AKR1C3 activity [ 73 ]. This SAR was expanded further with the filing of WO2025/214434, which saw the introduction of 42 macrocyclic derivatives [ 70 ]. Of these derivatives, 9 compounds reported a growth inhibition IC 50 <10 nM in H460 cells, with activity attenuated upon ASP9521 pretreatment ( Figure 10 ). In addition, Ascentawits Pharmaceuticals filed WO2023/226959 which investigated the use of AST-3424 in combination with various cell cycle inhibitors [ 74 ]. Within this patent, the inventors dosed HT-29 (colorectal adenocarcinoma) cells with AST-3424 for 72 hours, following a 24-hour pretreatment of adavosertib at a range of concentrations (100-220 nM). The inventors concluded that when dosed in combination, including at low doses of adavosertib (100 nM) greater cytotoxic effects were observed than when dosed individually, with a combination HT-29 growth inhibition IC 50 of 2.11 nM. The same trend was observed when dosed in combination with AZD7726 (100-220 nM) and palbociclib (50-10,000 nM) in HT-29 cells. The inventors also report the combination of AST-3424 with ceralasertib and report significant increase in both HT-29 and H490 cell growth inhibition IC 50 values as compared to AST-3424 alone. In HT-29 cells, AST-3424 has an IC 50 of 128.60 nM towards cell growth inhibition, but when dosed in combination with 1.3 μM of ceralasertib, the IC 50 value decreased to 1.42 nM. In H490 cells, the growth inhibition IC 50 improved from 0.40 nM with AST-3424 alone to 0.02 nM when dosed with 0.44 μM ceralasertib. In addition to targeting AKR1C3 directly, or compound activation by AKR1C3, two other patents have been filed that utilize AKR1C3 for disease detection and monitoring. WO2023/153998 investigates the relationship between AKR1C3 expression and prostate specific antigen (PSA) levels to determine risks associated with disease progression levels [ 75 ]. The inventors report that high AKR1C3 expression is associated with a significantly shorter overall survival time for patients diagnosed with CRPC. For patients expressing low levels of AKR1C3, only 1 of 15 patients died during their study, in comparison to 11 of 15 that expressed high AKR1C3 levels. Ultimately, by dividing patients into risk groups by using their derived receiver operating characteristic curve, the AKR1C3 levels could be used to help predict which patients may suffer shorter survival times. In addition, the inventors found correlation between AKR1C3 expression levels and successful abiraterone therapy. Of 25 CRPC patients, AKR1C3 was used as a positive predictive biomarker and identified 28% of patients that responded well to abiraterone treatment. This study revealed that through the monitoring of AKR1C3, patients expressing higher PSA levels, but low AKR1C3 levels, present a 3-fold increase in progression free survival, as compared to patients with high AKR1C3 expression. The utilization of AKR1C3 as a biomarker would prove invaluable upon confirmation of diagnostic reliability, as not only would healthcare providers have an alternative approach for monitoring disease progression, but this will also allow for monitoring response to chemotherapy in clinical trials. This will allow the option for personalized therapy and ensure that patients receive the most appropriate treatment at the time of diagnoses, conceivably minimizing risk of disease progression and improving quality of life. Ascentawits Pharmaceuticals filed, US2024/0142456 claiming the development of a diagnostic kit capable of detecting AKR1C3 [ 76 ]. Current immunohistochemical (IHC) methods are unable to achieve staining detection of AKR1C3 across various cancer tissues in a single IHC assay. As many treatment regimens used in the clinic are often based on the results observed from IHC staining, it is imperative that novel AKR1C3 detection methods must have excellent sensitivity, precision and must provide consistent results. The novel diagnostic kit introduces specific monoclonal antibodies for AKR1C3 as well as specific antigen retrieval conditions to obtain AKR1C3 from patient tissues, which had not been reported specifically towards AKR1C3 prior. Reproducibility was confirmed with two independent pathologists utilizing the kit on 46 samples, with 100% consistency observed between both pathologists. Ensuring reproducibility and consistency between pathologists is essential as real-world applications would see the diagnostic kit being used to different standards, including personal and environmental influences potentially affecting the results. Therefore, developing a robust and reliable method for monitoring AKR1C3 expression will ensure that obtained results are accurate and will prevent risk of misdiagnoses, which is a common issue observed when utilizing PSA levels as a biomarker for PC progression [ 77 ] This diagnostic kit will allow AKR1C3 overexpression to be used as a consistent biomarker when considering the development of androgen-dependent cancers and can be used as an alternative approach for disease progression monitoring, in which other methods, such as PSA monitoring, may not be sufficient.

For this review article, patents, and their related patent families, were obtained through the World Intellectual Property Organization (WIPO) database and SciFinder, accessed between October 2025 and March 2026. Journal articles were obtained through various databases, including PubMed, ScienceDirect and Scholar, as well as through the journals directly.

Aldo-keto reductase 1C3 (AKR1C3) is a key member of the AKR1C superfamily[ 1 ]. AKR1C3 has been commonly identified as a potential drug target due to its involvement in the development of many cancers, including endometrial[ 2 ], prostate[ 3 ], breast[ 4 ], and hematological malignancies[ 5 ] [ 6 ] [ 7 ], as well as other non-cancer hormonal disorders, such as endometrioses [ 2 ] and polycystic ovary syndrome (PCOS) [ 8 ]. Thus, AKR1C3 has become a highly-sought target for treating these diseases, with over 15 inhibitor series reported since 2021. [ 9 ] [ 10 ] [ 11 ] [ 12 ] [ 13 ] [ 14 ] Inhibitors of AKR1C3 have most commonly been explored for their therapeutic effect in castration-resistant prostate cancer (CRPC) and therefore, this review will predominantly focus on this malignancy. Despite low levels of circulating testosterone in patients post androgen deprivation therapy (ADT), the tumor continues to proliferate, which is attributed to the upregulation of additional mechanisms allowing for androgen production post-castration. The major mechanism is intratumoral steroidogenesis, in which production of intratumoral androgens can sustain androgen receptor (AR) signaling, facilitating tumor growth[ 15 ] [ 16 ]. The protein AKR1C3 plays a vital role in this mechanism, by catalyzing the conversion of androgen precursors into their more potent counterparts ( Figure 1 ), such as 5α-adione into 5α-dihydrotestosterone (5α-DHT), Δ 4 -androstene-3,17-dione into testosterone, and androsterone into 5α-andros-tane-3,17β-diol [ 2 ]. Abiraterone acetate, a prodrug of abiraterone, is clinically employed as a therapeutic approach to prevent the production of intratumoral androgens through the inhibition of P45017A1 (CYP17A1) [ 17 ] [ 18 ] [ 19 ]. Additionally, AKR1C3 has been shown to complex with the AR splice variant 7 (AR-v7) which promotes resistance to AR signaling inhibitors (ARSIs), including enzalutamide, and facilitate cell proliferation[ 20 ] [ 21 ] [ 22 ]. Enzalutamide works by impairing AR ligand binding and reducing the efficiency of its nuclear translocation and is the current clinical gold standard for treating CRPC. Despite these FDA approved treatments, approximately 25% of patients develop resistance to enzalutamide and approximately 33% of patients develop resistance to abiraterone [ 23 ]. Additional ARSIs are available, including apalutamide and darolutamide, however, resistance is still observed [ 24 ] [ 25 ]. Thus, there is a critical unmet need for the development of novel agents to overcome drug resistance. Studies have previously demonstrated that inhibition of AKR1C3 with indomethacin has shown re-sensitization of relevant xenograft models to enzalutamide [ 26 ]. In addition, treatment of enzalutamide resistant 22Rv1 cells (CRPC) with an AKR1C3 degrader has also shown re-sensitization to enzalutamide treatment [ 27 ], highlighting that further development of AKR1C3 inhibitors and/or degraders could be beneficial in surmounting resistance to ARSIs and slowing tumor progression. In the development of endometrial cancer, it has been shown that elevated levels of AKR1C3 increases the concentration of estradiol, resulting in increased proliferation of the endometrium [ 28 ]. Furthermore, AKR1C3 converts prostaglandin (PG)D 2 to 9α,11β-PGF 2 , preventing its eventual conversion to 15Δ-PGJ 2 , a natural ligand for the peroxisome proliferator-activated receptor-γ. In doing so, AKR1C3 provides activation of the mitogen activated protein kinase cascade, promoting aggressive cell proliferation and anti-apoptotic effect [ 29 ]. Hence, identification of AKR1C3 inhibitors may provide an alternative therapeutic approach to treating endometrial cancer [ 30 ]. Additionally, AKR1C3 plays a role in the growth of hormone-dependent breast cancer [ 31 ]. By acting as a local source of testosterone, AKR1C3 enables the production of 17β-estradiol by aromatase, driving cell proliferation [ 32 ] [ 33 ]. Aromatase inhibitors are a common treatment for ER+ breast cancer, however, 20% of patients with early-stage breast cancer develop resistance to aromatase inhibitors and resistance is considered inevitable in patients with metastatic disease [ 34 ] [ 35 ] [ 36 ]. Anthracyclines, such as doxorubicin and daunorubicin, are also commonly used to treat metastatic breast cancer. Studies have proposed that anthracycline resistance can be attributed to the reduction of anthracyclines into their respective alcohol metabolites, via AKR1C3, decreasing their activity and increasing their cardiotoxicity [ 37 ] [ 38 ] [ 39 ]. Hence, small molecule inhibitors of AKR1C3 show promise in preclinical models to provide therapeutic relief for patients with ER+ breast cancer who do not respond to treatment with aromatase inhibitors or anthracyclines. Development of AKR1C3 inhibitors remains a consistent challenge, due to the common morphology between the AKR superfamily, in which AKR1C3 shares greater than an 86% sequence morphology with AKR1C1, AKR1C2 and AKR1C4 [ 27 ]. In addition, these members play critical roles in the prevention of various cancer progression where 5α-DHT is metabolized to lesser active 3α-diol by AKR1C2 and 3β-diol by AKR1C1, while AKR1C4 contributes to the formation of bile acids, and thus inhibition must be avoided [ 40 ] [ 41 ] [ 42 ]. Despite this, 13 patents have been filed since 2020, including various AKR1C3 inhibitors and a first-in-class degrader, as well as patents that utilize AKR1C3 to metabolically cleave prodrug moieties, as potential therapeutic alternatives for various diseases. An excellent patent review on AKR1C3 inhibitors was published in 2017, covering the development of various steroidal and non-steroidal AKR1C3 inhibitors that had been claimed up to 2017 . This review aims to expand on this prior patent review by expanding coverage to novel AKR1C3 inhibitors claimed between 2020 to present, as well as including insight into the emerging areas of AKR1C3-mediated prodrug activation and potential use of the protein as a diagnostic biomarker.