Reverse repurposing: Potential utility of cancer drugs in nonmalignant illnesses

Intriguingly, a significant number of molecular pathways are altered in both non-malignant and malignant conditions, often with the exact same gene mutation being responsible for both types of conditions. Moreover, several drugs or drug classes now carry approvals for both benign and malignant conditions. Examples include but are not limited to MEK and PI3K/AKT/mTOR pathway inhibitors, FGFR and JAK2 inhibitors, Bruton’s tyrosine kinase inhibitors, and CD20-directed antibodies (CD20 being a B-cell surface marker). Importantly, many non-malignant but serious illnesses, including but not limited to Alzheimer’s and other neurodegenerative diseases, endometriosis, dwarfism syndromes such as achondroplasia, and autoimmune/inflammatory conditions, harbor molecular pathway aberrations or even specific genomic mutations implicated in cancer. Several examples are shown in Figure 2 and Figure 3 . There are a number of both druggable and not yet druggable driver genes found altered in malignancy that also have similar or indistinguishable alterations which characterize benign, premalignant, hereditary germline conditions, and/or somatic mosaic conditions 17 , 18 . Why some of these non-malignant conditions do not develop into malignancy is a fascinating question, which remains for the most part unanswered. Even so, targeted therapeutics used in oncology have potential to treat these conditions. Table 1 provides examples of targeted therapeutics with FDA-approvals, clinical data, or with theoretical value for non-malignant conditions with mutations in oncogenic drivers. The RAF-MEK-ERK signaling cascade is a well-characterized pathway involved in cell proliferation and survival. Alterations of this pathway are found in over one-third of cancers and are considered strong drivers of oncogenesis 72 . It is therefore rather unexpected that identical gene anomalies are found in many different non-malignant conditions. As an example, BRAF alterations occur in approximately 15% of cancers 73 and they can be effectively targeted across most cancers with approved BRAF (e.g., vemurafenib, dabrafenib, encorafenib) and/or MEK inhibitors (e.g., trametinib, cobimetinib, binimetinib). Indeed, the combination of the BRAF inhibitor dabrafenib and the MEK inhibitor trametinib has received a tissue-agnostic FDA approval for solid cancers harboring BRAF V600E mutations 74 . Paradoxically, activating, deleterious BRAF mutations (i.e., BRAF V600E ) occur in ~75% of benign melanocytic proliferations such as nevi despite the near complete lack of evolution of these benign nevi into melanomas; moreover, BRAF V600E , a clearly targetable driver susceptible to FDA-approved BRAF or MEK inhibitors, occurs in only 50-60% of melanoma, hence less than in its benign counterpart 75 , 76 . There are case reports of involution of benign nevi after MEK pathway therapy (e.g., with cobimetinib) 77 , though this may not be a common occurrence. NRAS alterations and BRAF -activating alterations including fusions have also been seen in giant congenital melanocytic nevi 78 and may be treatable with MEK inhibitors 52 , 54 . The RASopathies are a group of inherited syndromes that are caused by germline pathogenic alterations in genes encoding the RAS/MEK signaling pathway 79 . As an example, Noonan syndrome is an autosomal dominant disorder characterized by widely set eyes, low-set ears, short stature, and pulmonic stenosis; the gene defect is usually in PTPN11 also known as SHP2 . PTPN11/SHP2 mediates RAS/MEK/ERK activation through Grb2-associated binder protein. Cardio-facio-cutaneous syndrome and Costello syndrome are other examples of RASopathies. For unknown reasons, RASopathy-associated cancers are usually of different histological origin than those seen with sporadic mutations of the same genes 80 . The preclinical evaluation of targeted agents specific for the RAS/MAPK pathway in models of RASopathies suggest that MEK inhibitors show therapeutic promise 28 , 29 , 79 . Furthermore, clinical successes have been reported. For instance, there are reports or reversal of severe hypertrophic cardiomyopathy and chylous effusions in children with Noonan syndrome treated with the MEK inhibitor trametinib 44 - 46 . Somatic KRAS alterations have been reported in AVM of the brain (a leading cause of hemorrhagic stroke in young people) 42 . Case reports demonstrating successful treatment of AVM with the MEK inhibitor trametinib have been published 4 , 5 and preclinical mouse models support trametinib as a therapeutic option 43 . There is theoretical benefit from MEK inhibitors for treatment of endometriosis. Anglesio and colleagues found that lesions in deep infiltrating endometriosis, which is associated with virtually no risk of malignant transformation, harbor somatic cancer driver mutations, including in PIK3A and KRAS genes 81 . The resultant intensification of kinase signaling represents a viable target for repositioning cancer drugs that suppress the PI3K/AKT/mTOR or MEK axis in order to treat endometriosis 82 . Alzheimer’s disease is a progressive neurodegenerative disorder and the most common cause of dementia worldwide, with no adequate treatment. The prevalence of Alzheimer’s disease is increasing dramatically due to the aging of the world population. Pathogenic brain somatic mutations identified in a subset of individuals suffering from Alzheimer’s disease are enriched in PI3K-AKT and MEK pathway gene aberrations known to contribute to hyperphosphorylation of tau, a microtubule-associated protein that accumulates as neurofibrillary tangles in Alzheimer’s disease 83 . Hence there is growing interest in these pathways as therapeutic targets for Alzheimer’s disease. Neurofibromatosis type 1 is a hereditary disorder characterized by NF1 alterations, which activate the MEK pathway; selumetinib is a MEK inhibitor that is FDA-approved for treatment of people with neurofibromatosis type 1 (NF1) who have symptomatic, inoperable plexiform neurofibromas 84 and clinical data is also present for use of trametinib in this disorder 48 . ARAF recurrent mutations can cause central conducting lymphatic anomaly; in a child, treatment with a MEK inhibitor resulted in remodeling of the patient's lymphatic system with resolution of the lymphatic edema, as well as marked improvement in his pulmonary function tests, with cessation of supplemental oxygen requirements and near normalization of daily activities 85 . Genes in the PI3K/AKT/mTOR pathway are frequently aberrant in a wide range of cancers, being discerned in almost two-fifths of malignant neoplasms 6 . The pathway can be targeted by multiple approved cancer drugs, including but not limited to alpelisib (PIK3CA inhibitor), capivasertib (AKT inhibitor) and nab-sirolimus, everolimus and temsirolimus (mTOR inhibitors) 7 . There are also diverse non-malignant conditions that harbor alterations in this pathway, with the potential for repositioning of cancer drugs. Several inherited or mosaic conditions are due to alterations in the PI3K/AKT/mTOR pathway. As an example, tuberous sclerosis complex is an autosomal dominant disorder characterized by skin manifestations and formation of multiple tumors in different organs, mainly in the central nervous system. Tuberous sclerosis is caused by the mutation of one of two tumor suppressor genes-- TSC1 or TSC2 , which in turn leads to activation of the mTOR pathway 86 . It is fascinating that the FDA has approved everolimus, an mTOR inhibitor authorized for cancer, for tuberous sclerosis complex-associated partial-onset seizures 87 . PTEN is altered in hamartoma tumor syndrome 55 (includes Cowden’s disease), an autosomal dominant genodermatosis in which non-cancerous hamartomas develop in different body areas. In this disorder, PTEN germline alterations generally result in PTEN loss of function that secondarily upregulates the PI3K/AKT/mTOR axis. Intriguingly, again, there is improvement including in neurocognitive symptoms following everolimus treatment 56 . These results indicate that matched molecular targeting can ameliorate non-malignant sequelae of gene defects, including neurologic impairment. Neurofibromatosis-2 is characterized by germline NF2 mutations, which in turn activate the PI3K/AKT/mTOR axis 51 ; there is clinical data supporting the use of temsirolimus for this condition 49 . Peutz-Jegher syndrome is a hereditary syndrome characterized by gastrointestinal polyposis, mucocutaneous pigmented macules, and predisposition to certain cancers. It is caused by germline STK11 mutations, which activate the mTOR axis, potentially suggesting that mTOR suppression might be a treatment option that merits study 58 , 88 . Proteus syndrome is generally characterized by a mosaic AKT1 pathogenic variant (not germline) with progressive patchy overgrowth of the skeleton, skin, adipose, and central nervous systems, causing severe disfigurement, starting from the toddler stage, and predisposition to deep vein thrombosis, pulmonary embolism and a range of tumors 89 . There is clinical data for the benefit for the AKT inhibitor miransertib for proteus syndrome 23 . PI3K pathway alterations have also been observed activated PI3K delta syndrome (APDS) (autosomal dominant inheritance), and PIK3CA-related overgrowth spectrum (PROS), a heterozygous mosaic (or rarely, constitutional) syndrome due to an activating deleterious PIK3CA mutation 9 , 18 , 71 . Alpelisib is approved for PROS 90 and leniolisib (selective PI3Kδ inhibitor) is approved for APDS 91 , which is caused by mutations in PIK3CD or PIK3R1 genes that encode the PI3Kdelta protein. PIK3CA alterations have also been observed in acquired benign conditions such as seborrheic keratosis, endometriosis, and Alzheimer’s disease. Seborrheic keratosis, though benign and without malignant potential, frequently display the typical cancer-associated PIK3CA mutations E542K, E545K, and H1047R 92 . SM-020 is a selective and potent AKT inhibitor applied by patients at home to their seborrheic keratoses lesions with preliminary positive results 93 . In endometriosis, in addition to alterations in the MEK pathway, as discussed above, alterations in the PI3K/AKT/mTOR pathway are also common. A number of pharmaceutical compounds that target these pathways have been successfully applied in preclinical models of endometriosis 82 . Finally, in Alzheimer disease, in addition to brain somatic mutations identified in the MEK pathway, the PI3K-AKT pathway genes may also be mutated and participate in Alzheimer’s pathogenesis, perhaps through hyperphosphorylation of tau 83 . Fibroblast growth factor receptors (FGFRs) are aberrantly activated through single-nucleotide variants, gene fusions and copy number amplifications in 5–10% of human cancers 94 . Several drugs with FGFR inhibitory activity, including but not limited to erdafitinib, pemigatinib, futibatinib and infigratinib (recently withdrawn) achieved FDA approval for cancers with cognate FGFR aberrations 94 . In addition, other multikinase inhibitors such as lenvatinib (FDA approved for several types of cancer, but approval not based on an FGFR or other biomarker), have robust FGFR inhibitory activity. Importantly, FGFR alterations also drive an assortment of inherited and acquired non-malignant illnesses. Hereditary FGFR alterations have been observed in a variety of hereditary syndromes, e.g., Hartsfield syndrome (autosomal dominant, FGFR1 loss-of-function alterations), Kallman syndrome (sometimes have loss-of-function FGFR1 alterations), and Pfeiffer syndrome (gain-of-function FGFR1 or FGFR2 alterations, autosomal dominant) 95 . Pfeiffer syndrome is now known to be a member of a group of conditions caused by variants in the FGFR genes including Apert syndrome, Crouzon syndrome, Beare-Stevenson syndrome, FGFR2 -related isolated coronal synostosis, Jackson-Weiss syndrome (all craniosynostosis syndromes with FGFR2 gain-of-function hereditary anomalies) 96 . Muenke syndrome is a craniosynostosis characterized by FGFR3 gain-of-function mutations 97 . FGFR inhibitors may theoretically be of benefit in syndromes with gain-of-function FGFR alterations, but not in syndromes with loss-of-function FGFR alterations 31 . Of special interest, achondroplasia is the most common form of human dwarfism; it is caused by germline activating FGFR3 mutations; significantly, though the FGFR3 mutations can be indistinguishable from those found as targetable somatic mutations in some human malignancies such as bladder cancers, people with achondroplastic dwarfism do not have an increased incidence of cancer 98 . Infigratinib, a potent pan-FGFR inhibitor, is being investigated in children with achondroplastic dwarfism, since FGFR inhibitors demonstrated the ability to reverse the achondroplastic phenotype in mouse models 32 , 33 , 99 . In achondroplastic children, early results suggest that FGFR inhibitor therapy significantly increases height velocity 33 , 100 . A question that will need to be answered is whether there is a change in susceptibility to cancer in people who have pathogenic FGFR3 germline mutations and pharmacological suppression from childhood, especially should they stop the drug. Neuregulins are a family (NRG1 (types I–VI), NRG2, NRG3, and NRG4) of EGF-like signaling molecules involved in the development and repair of diverse body elements including those of the nervous system, skeletal muscle, heart, breast, and other organs. NRG1 has undergone significant study; it binds to ERBB3/HER3 and ERBB4/HER4 and also influences EGFR/ERBB1/HER1 and ERBB2/HER2 receptor signaling via heterodimerization with ERBB3/HER3 and ERBB4/HER4. NRG1 -activating fusions can be found in diverse cancer types, albeit at a low rate — ~0.15–0.5% across tumors 101 . Of interest, there is growing evidence that elevated NRG1 expression may be implicated in the pathogenesis of amyotrophic lateral sclerosis (ALS), Alzheimer’s disease and schizophrenia. In ALS, it has been postulated that NRG1 upregulation may occur in response to ERBB4 loss-of-function mutations, which may in turn then overstimulate ERBB1, ERBB2 and ERBB3 signaling 30 . It is not known if drugs impacting NRG1 or its downstream effectors, such as the pan-ERBB inhibitors afatinib, dacomitinib or neratinib, which are approved for various cancers, could impact the neurodegenerative and psychiatric diseases in which NRG1 appears to be overexpressed. Multiple molecular targets are utilized for malignant and non-malignant indications. Inhibitors of Bruton’s tyrosine kinase (BTK), CD20, fibroblast growth factor receptor (FGFR), phosphatidylinositol 3-kinase (PI3K), MEK, interleukin-6 (IL-6), mammalian target of rapamycin (mTOR), and Janus kinase (JAK), along with gonadotropin-releasing hormone (GnRH) agonist/antagonist therapies are summarized in Tables 2 and 3 and Figure 2 . The mode of action of these therapies in non-malignant conditions are summarized in Table 2 . Autoimmune conditions can have abnormal B-cell activation that is driven by BTK and CD20 activation. T lymphocyte activation and proliferation leading to rejection can occur through the mTOR pathway. Chronic inflammatory and autoimmune conditions may have IL-6 and JAK pathway activation. Abnormal cell proliferation can be driven by the PI3K, RAS, and FGFR pathways. Thus, specific inhibitors of these pathways are approved and of benefit in a number of benign conditions. There are four BTK inhibitors approved for malignant indications. Acalabrutinib, ibrutinib, and zanubrutinib are all authorized for chronic lymphocytic leukemia (CLL) and small lymphocytic lymphoma (SLL). Additionally, acalabrutinib is approved for mantle cell lymphoma (MCL), ibrutinib for Waldenström macroglobulinemia, and zanubrutinib for MCL, marginal zone lymphoma (MZL), and Waldenström macroglobulinemia. Pirtobrutinib is approved for MCL. Ibrutinib also carries a non-malignant indication for chronic graft-versus-host disease (GVHD). BTK is hyperactivated on B-cells in chronic GVHD, and thus inhibition with ibrutinib can effectively treat this non-malignant disease 64 , 102 . Four CD20-directed antibodies are approved for malignant indications. Obinutuzumab, ofatumumab, and rituximab are approved for CLL. Obinutuzumab is also approved for follicular lymphoma (FL) and rituximab for non-Hodgkin’s lymphomas (NHL). Ibritumomab tiuxetan is a CD20-directed radiotherapeutic antibody approved in NHL 103 . There are four CD20-directed antibodies approved for non-malignant indications; ocrelizumab, ofatumumab, and ublituximab are all approved for use in multiple sclerosis (MS), whereas rituximab carries indications for granulomatosis with polyangiitis (GPA), microscopic polyangiitis, pemphigus vulgaris, and rheumatoid arthritis. CD20-directed antibodies lead to the depletion of the pathogenic, self-reactive B-cells that drive these autoimmune diseases 65 , 104 . The potential for targeting FGFR gain-of-function mutations that characterize non-malignant disease by repositioning FGFR inhibitors approved for cancer with FGFR activating fusion/alterations/mutations has been discussed (section FGFR alterations and inhibitors and in Table 1 ). In addition to targeting FGFR mutations, fibrosis may be causative for specific disorders, and FGFR inhibitors may show activity in these conditions. For instance, nintedanib is an FGFR1-3 inhibitor that is approved for several pulmonary indications including idiopathic pulmonary fibrosis, progressive chronic fibrosing interstitial lung disease (ILD), and systemic sclerosis-associated ILD 105 . Nintedanib competitively inhibits multiple tyrosine kinases, including FGFR1-3, blocking intracellular signaling cascades that are involved in the pathogenesis of fibrotic tissue remodeling in ILD fibrosis, including fibroblast proliferation, migration and differentiation 106 . Four PI3K pathway inhibitors have been approved for oncology indications. Alpelisib is approved for PIK3CA -mutated breast cancer. Copanlisib was approved for use in follicular lymphoma, however as of November 2023 it is being voluntarily withdrawn from the market due to lack of efficacy seen in the confirmatory phase 3 trial. Duvelisib and idelalisib are both approved for CLL. Alpelisib and a distinct PI3K-delta inhibitor, leniolisib, are approved for non-malignant indications. Alpelisib inhibits the PI3Kα pathway, preventing downstream overgrowths and malformations comprising a wide group of clinically recognizable disorders commonly known as PIK3CA -related overgrowth spectrum (PROS). Interestingly, alpelisib may also attenuate excessive insulin production in congenital hyperinsulinemia through modulation of the PI3K pathway 107 . Leniolisib inhibits the PI3K delta pathway, preventing production of phosphatidylinositol (3,4,5)-trisphosphate (PIP3), hyperactivity of the downstream mTOR/AKT pathway, and the proliferation and activation of B and T cell subsets seen in activated PI3K delta syndrome (APDS) 108 - 110 . There are three MEK inhibitors with oncology indications. Binimetinib and cobimetinib are approved for melanoma with BRAF V600E or V600K alterations. Binimetinib is also approved for non-small cell lung cancer (NSCLC) with a BRAF V600E mutation, and cobimetinib for histiocytic neoplasms. There are multiple cancers with approvals for trametinib in BRAF V600E or V600K mutated tumors and, more recently, trametinib received a tissue-agnostic approval when given in combination with dabrafenib for solid tumors with BRAF V600E mutations 111 . Selumetinib is a distinct MEK inhibitor approved for neurofibromatosis type 1. Selumetinib inhibits MEK1/2, preventing downstream phosphorylation of ERK, leading to reduced neurofibroma numbers, volume, and proliferation 112 , 113 . (see also section on BRAF/MEK pathway alterations and BRAF/MEK inhibitors ) The IL-6 inhibitor siltuximab is approved for Castleman disease, a lymphoproliferative disorder whose hallmark is elevated IL-6 levels 63 . Whether Castleman disease should be considered an autoimmune disease or cancer, is currently unclear, though in patients with multicentric disease, the behavior can be similar to cancer 114 . There are three distinct IL-6 antibodies--satralizumab, sarilumab, and tocilizumab--that have non-malignant indications. Satralizumab is approved for neuromyelitis optica spectrum disorder. Sarilumab is approved for polymyalgia rheumatica and rheumatoid arthritis. Tocilizumab is approved for chimeric antigen receptor (CAR) T-cell-associated cytokine release syndrome, COVID-19 treatment, giant cell arteritis, rheumatoid arthritis, systemic juvenile idiopathic arthritis, and systemic sclerosis-associated ILD. Satralizumab, sarilumab, and tocilizumab bind to both soluble and membrane-bound IL-6 receptors, inhibiting IL-6-mediated signaling through these receptors; IL-6 is involved in T-cell activation, induction of immunoglobulin secretion, initiation of hepatic acute phase protein synthesis, and stimulation of hematopoietic precursor cell proliferation and differentiation. IL-6 is also produced by synovial and endothelial cells leading to local production of IL-6 in joints affected by inflammatory processes such as rheumatoid arthritis 115 . Several mTOR inhibitors are approved for oncology indications: everolimus is approved for breast cancer, neuroendocrine tumors, renal cell carcinoma, and tuberous sclerosis complex (TSC)-associated subependymal giant cell astrocytoma; nab-sirolimus, for perivascular epithelioid cell tumor and temsirolimus is approved for renal cell carcinoma 116 ; and temsirolimus is approved for kidney cancer. Everolimus also has multiple non-malignant indications: liver transplant, renal transplant, TSC-associated partial-onset seizures, and TSC-associated renal angiomyolipoma. Sirolimus, another mTOR inhibitor, is approved for facial angiofibroma associated with TSC, kidney transplant, and lymphangioleiomyomatosis 116 (see also section on PI3K/AKT/mTOR pathway alterations and inhibitors ). There are four JAK inhibitors approved for oncology indications. Pacritinib, momelotinib, ruxolitinib, and fedratinib are all approved for myelofibrosis. Ruxolitinib has an additional approval for polycythemia vera. Somatic gain-of function alterations in various JAK genes are predominantly associated with hematologic malignancies 117 . There are also multiple JAK inhibitors approved for non-malignant indications 67 . JAK pathway activation can lead to autoimmunity through T-cell differentiation, lymphocyte effector function, and macrophage activation 67 ; thus JAK2 inhibitors can benefit many autoimmune conditions. Abrocitinib is approved for atopic dermatitis. Barictinib is approved for rheumatoid arthritis, alopecia areata, and COVID-19. Deucravacitinib is approved for plaque psoriasis, ritlecitinib is approved for alopecia areata, and ruxolitinib is approved for GVHD. The topical formulation of ruxolitinib is also approved for atopic dermatitis and nonsegmental vitiligo. There are multiple approvals for tofacitinib, which include rheumatoid arthritis, psoriatic arthritis, ulcerative colitis, juvenile arthritis, and ankylosing spondylitis. Upadacitinib has approvals for rheumatoid arthritis, psoriatic arthritis, atopic dermatitis, Crohn disease, ulcerative colitis, ankylosing spondylitis, and spondyloarthritis 67 . GnRH agonists and antagonists are short peptide analogues of GnRH that suppress estrogen and androgen synthesis and are often used as androgen deprivation therapy in patients with advanced prostate cancer. Some of these agents are also administered to manage benign conditions responsive to hormonal inhibition such as endometriosis, uterine fibroids, precocious puberty and infertility 118 . Goserelin, histrelin, leuprolide, and triptorelin are GnRH agonists approved for prostate cancer. Goserelin has an additional approval in breast cancer. Five GnRH agonists also carry approvals in non-malignant conditions. Goserelin, nafarelin, and leuprolide are each approved in endometriosis. Histrelin, leuprolide, and triptorelin are each approved for central precocious puberty. Goserelin also has an approval for endometrial thinning agent prior to endometrial ablation for dysfunctional uterine bleeding, and leuprolide has an approval for uterine fibroids. Degarelix and relugolix are GnRH antagonists approved for prostate cancer. Cetrorelix and ganirelix are GnRH antagonists approved for controlled ovarian stimulation, while elagolix is approved for endometriosis. Cetrorelix, elagolix, and ganirelix competitively inhibit the GnRH receptors on the pituitary and induce rapid, reversible suppression of gonadotropin secretion. Sustained dosing of elagolix leads to a decrease in serum estradiol to postmenopausal levels, such that endometrial tissue that depends on gonadal steroids for maintenance becomes quiescent. Short-term dosing of cetrorelix or ganirelix delays the LH-surge, and consequently ovulation, in a dose-dependent fashion until ovarian stimulation with hCG. We found 36 therapeutics that have both FDA-approved malignant and non-malignant indications. These are summarized in Table 3 . Original FDA indications are listed in Table S1 ( FDALabel [database online]: Drug Product Labeling. U.S Food & Drug Administration. https://nctr-crs.fda.gov/fdalabel/ui/search ). Examples of therapies that are FDA-approved for a malignant condition that have clinical data, but are not FDA-approved, are summarized in Table S2 . Cytotoxic chemotherapy blocks cell proliferation and growth in malignant conditions and can lead to immunosuppression through prevention of immune cell proliferation. Thus, many cytotoxic chemotherapeutics have shown benefit for non-malignant autoimmune conditions and those characterized by abnormal cell growth. 5-fluorouracil, a pyrimidine nucleoside analogue, was originally approved for colorectal cancer prior to receiving additional approvals for breast cancer, gastric cancer, pancreatic cancer, and superficial basal cell carcinoma. It is also approved for actinic keratosis. Cladribine is a purine nucleoside analog initially approved for hairy cell leukemia and then later approved for multiple sclerosis. Methotrexate, a folate antimetabolite, was originally approved for acute leukemia. It subsequently acquired approvals for breast cancer, head and neck cancer, lung cancer, meningeal leukemia, mycosis fungoides, non-Hodgkin’s lymphoma, osteosarcoma, and gestational trophoblastic neoplasia. It is used for the autoimmune conditions psoriasis, rheumatoid arthritis, and polyarticular juvenile idiopathic arthritis. Mitomycin is an anticancer antibiotic which was originally approved for gastric and pancreatic cancers. It is also used for urothelial cancers and as an adjunct to glaucoma surgery. Mitoxantrone is an anthracycline originally approved for acute myeloid leukemia. It is also approved for prostate cancer and multiple sclerosis. Hydroxyurea is an antimetabolite which had an original malignant approval in 1967. It is currently approved for chronic myelogenous leukemia and head and neck cancer along with sickle cell anemia. Alemtuzumab, a CD52-directed monoclonal antibody, was initially approved for B-cell chronic lymphocytic leukemia. It was later approved for multiple sclerosis. Daratumumab, a CD38-directed monoclonal antibody, was originally approved for multiple myeloma and was later approved for the non-malignant condition light chain amyloidosis. Ofatumumab, a CD20-directed monoclonal antibody was originally approved for CLL and was repurposed for multiple sclerosis. Rituximab, a CD20-directed monoclonal antibody, was originally approved for non-Hodgkin’s lymphoma. It was later approved for use in CLL. Non-malignant indications include granulomatosis with polyangiitis, microscopic polyangiitis, pemphigus vulgaris, and rheumatoid arthritis. Alpelisib, a PI3K inhibitor, was originally approved for breast cancer, then later received approval for PIK3CA -related overgrowth spectrum (PROS). Crizotinib is a multi-tyrosine kinase inhibitor originally approved for non-small cell lung cancer. It later received approvals for anaplastic large cell lymphoma and the non-malignant inflammatory myofibroblastic tumor. Everolimus is an mTOR inhibitor initially approved for renal cell carcinoma. It has additional malignant indications of breast cancer, pancreatic neuroendocrine tumor, and subependymal giant cell astrocytoma associated with tuberous sclerosis complex (TSC). It is used for several non-malignant indications which include liver transplant, renal angiomyolipoma with TSC, renal transplant, and TSC-associated partial-onset seizures. Ibrutinib, a Bruton’s tyrosine kinase inhibitor, was initially approved for mantle cell lymphoma and has subsequently received approvals for CLL/SLL and Waldenström macroglobulinemia. It is also approved for graft-versus-host disease. Ruxolitinib, a Janus kinase inhibitor, was initially approved for myelofibrosis then later polycythemia vera. It has been repurposed to treat atopic dermatitis, nonsegmental vitiligo, and graft-versus-host disease. Interferon alfa-2b, an interferon, was originally approved for hairy cell leukemia. It acquired additional approvals for follicular lymphoma, hairy cell leukemia, Kaposi sarcoma, and melanoma. It is used in the treatment of non-malignant conditions including condylomata acuminata, hepatitis B, and hepatitis C. The monopegylated conjugate, ropeginterferon alfa-2b, was also later approved for polycythemia vera. Leuprolide, a GnRH agonist, was originally approved for prostate cancer. It is also approved for endometriosis, uterine fibroids, and central precocious puberty. Triptorelin, another GnRH agonist, was similarly first approved for prostate cancer with a subsequent approval for central precocious puberty. Goserelin, a GnRH agonist, was initially approved for prostate cancer prior to acquiring a breast cancer indication. It is also approved for endometrial thinning agent prior to endometrial ablation for dysfunctional uterine bleeding, and endometriosis. Megestrol acetate is a progestin that was first approved for endometrial cancer. It acquired additional approvals for breast cancer and for AIDS-associated anorexia/cachexia. Octreotide is a somatostatin analog originally approved for carcinoid tumors and VIPomas. It is also used for acromegaly. There were nine therapies identified which were originally approved for a non-malignant indication and later received approval for a malignant indication ( FDALabel [database online]: Drug Product Labeling. U.S Food & Drug Administration. https://nctr-crs.fda.gov/fdalabel/ui/search ). All-trans retinoic acid is a retinoid originally approved for acne vulgaris and then later approved for treatment of acute promyelocytic leukemia. It is also approved for photoaging. Denosumab is a RANKL-directed monoclonal antibody which was originally approved for osteoporosis prior to approval for giant cell tumor of the bone. It is also approved to treat hypercalcemia of malignancy and for prevention of skeletal-related events. Eflornithine, an ornithine decarboxylase inhibitor, received its first approval for the infectious disease indication of West African trypanosomiasis with central nervous system (CNS) involvement, but was later also approved for hirsutism as a topical agent and then for high-risk neuroblastoma. The GnRH agonist histrelin received its original approval for central precocious puberty and was later approved for prostate cancer. Imiquimod is a toll-like receptor 7 agonist which was originally approved to treat genital warts. It was later approved for basal cell carcinoma and is also used for actinic keratosis. Lanreotide, a somatostatin analog, had its original approval for acromegaly. It was later approved for gastroenteropancreatic neuroendocrine tumors and carcinoid syndrome. Raloxifene, a selective estrogen receptor modulator, was originally approved for osteoporosis prevention. It was later approved for risk reduction for recurrence of invasive breast cancer. Sirolimus, an mTOR inhibitor, was originally approved for kidney transplant. It was later approved for malignant perivascular epithelioid cell tumor treatment. Additional non-malignant indications include facial angiofibroma associated with tuberous sclerosis and lymphangioleiomyomatosis. Thalidomide was originally approved for erythema nodosum leprosum, then later acquired approval for treatment of multiple myeloma. Of note there were six therapies which were approved for use prior to 1960 for which it was difficult to discern if the initial indication was malignant or non-malignant. Five of these therapies were corticosteroids. Prednisone, hydrocortisone, methylprednisolone, and prednisolone all have approved indications in leukemia and lymphoma treatment. They are all indicated in numerous conditions. The general classes of approved indications include allergic states, collagen diseases, dermatologic diseases, edematous states, endocrine disorders, gastrointestinal diseases, hematologic disorders, ophthalmic diseases, respiratory diseases, and rheumatic disorders. Dexamethasone is used for leukemia, lymphoma, and multiple myeloma. The non-malignant applications include allergic states, dermatologic diseases, endocrine disorders, gastrointestinal diseases, and hematologic disorders. Fluoxymesterone is a synthetic androgen approved for breast cancer, delayed puberty, and hypogonadism.

Drug repurposing is an important strategy to identify therapeutic agents from existing approved drugs or compounds in clinical trials and use them in a different disease. Since drug discovery is expensive and time-consuming, drug repurposing can dramatically accelerate the timeline for application of successful therapeutics. In order to repurpose medications effectively, a deep knowledge of underlying biology of various conditions and scientific bridging between specialties is critical. Because cancer is one of the major causes of morbidity and mortality across the globe, there has been an emphasis on drug repurposing for compounds developed for benign illnesses that might be exploitable to treat cancer 119 . Reverse repurposing of drugs from malignant disease to non-malignant conditions has received less attention and is hampered by limited cross-fertilization between fields. Perhaps unexpectedly, there are numerous non-malignant conditions that harbor genomic alterations that, in the cancer field, are known to be druggable oncogenic drivers. Although patients afflicted with some of the non-malignant conditions characterized by genomic aberrations indistinguishable from those found in cancer do have a predisposition to cancer, others do not. It is puzzling as to why an oncogenic driver can be present and cause disease, albeit without malignant transformation potential. Regardless, it is now apparent that a variety of inherited and acquired benign disorders share activation pathways with cancer, such as the MEK, PI3K/AKT/mTOR, JAK, FGFR, and NRG/ERBB signals, and even the same molecular alterations. Because some of these oncogenic drivers/pathways are targetable in cancer patients with highly effective cognate agents, the question arises as to whether these agents could be exploited to treat non-malignant conditions sharing activation pathways and/or harboring identical mutations/alterations. To date, such repurposing of drugs has proven effective for several conditions. For instance, the PIK3CA inhibitor alpelisib (approved for breast cancer) is effective in reversing the phenotypic manifestations of PIK3CA-related overgrowth spectrum (PROS) and may also ameliorate excessive insulin production in congenital hyperinsulinemia through modulation of the PI3K pathway. MEK alterations such as KRAS mutations (common in cancer) are found in AVMs, and MEK inhibitors such as trametinib can anecdotally attenuate the impact of these AVMs. Everolimus, an mTOR inhibitor, is approved for alleviating partial seizures that afflict patients with tuberous sclerosis, a disease whose molecular hallmark is inherited TSC mutations that upregulate the mTOR pathway. Everolimus can also improve neurocognitive function in the PTEN hamartoma syndrome, a congenital syndrome due to germline PTEN loss-of-function alterations that result in upregulation of the mTOR signal. While there is great potential for repurposing oncology therapeutics to non-malignant conditions, there are several potential challenges to consider. Targeting molecular alterations in the non-malignant condition may have reduced efficacy compared to the malignant condition. The threshold to administer a therapy with side effects may be higher in non-malignant conditions as compared to metastatic cancer. The high cost of anti-neoplastic agents may be limiting in patients with non-malignant conditions who need to remain on continued therapy for years to decades. It is also unclear if ongoing therapy in non-malignant conditions will predispose to therapeutic resistance that will make these conditions harder to treat after initial therapy. Whether or not the use of targeted drugs that suppress the biologic impact of specific oncogenic mutations in conditions that have a cancer predisposition would also decrease the likelihood of developing cancer is not known. Importantly, many non-malignant but serious illnesses, including but not limited to Alzheimer’s disease and other neurodegenerative disorders, endometriosis, dwarfism syndromes such as achondroplasia, and autoimmune/inflammatory conditions, harbor pharmacologically tractable molecular pathway aberrations or even specific genomic mutations implicated in cancer.

Aberrant proliferative and immune processes can result in the development of cancer and can also lead to non-malignant conditions. Surprisingly, molecular interrogation of tissue affected by non-malignant illnesses, which may be congenital or acquired, and pre-malignant or not associated with malignancy, at times uncover “oncogenic” driver genomic alterations. Oncogenic drivers are alterations in genes that encode signals crucial for maintaining normal cellular growth and survival; these alterations have a confirmed role in the initiation and/or progression of cancer. Importantly, some of these oncogenic driver aberrations are now druggable in the context of cancer, often with impressive responses. The natural question that arises is whether drugs developed for targeting specific molecular abnormalities in cancer could be repurposed to treat non-malignant illnesses that harbor the same or similar molecular aberrations. Although there remains a lack of adequate cross-fertilization between fields, some successes in reverse repurposing, that is repurposing drugs from the cancer field to non-malignant conditions, have already been reported. For instance, MEK pathway activation due to somatic mutations in KRAS or other MEK pathway genes is frequently found in malignancy, being discerned in over 30% of tumors 1 , 2 . Notably, excessive activation of the MEK pathway has also been observed in arteriovenous malformations (AVM), including those in the brain, a non-malignant, but difficult-to-manage condition; this activation can be due to somatic activating KRAS mutations, though other genes in the pathway may also be aberrant 3 . Intriguingly, Lekwuttikarn et al reported that, in a child with an extracranial AVM and a somatic MAP2K1 mutation (also known to activate the MEK pathway), administration of the MEK inhibitor trametinib, which is approved for cancers such as melanoma harboring MEK pathway aberrations, resulted in reduced size of the malformation and lightening of color 4 . Similarly, Nicholson and colleagues reported successful management of an AVM with trametinib in a teenager with capillary-malformation AVM syndrome and cardiac compromise 5 . Other important examples of repurposing drugs relate to medications targeting PI3K/AKT/mTOR pathway aberrations. Genes in this pathway are frequently mutated in cancer, being seen in ~38% of malignancies 6 . There are several drugs that are approved by the Food and Drug Administration (FDA) for various malignancies, including PIK3CA inhibitors (e.g., alpelisib for breast cancer), AKT inhibitors (e.g., capivasertib for breast cancer) and mTOR inhibitors (e.g., everolimus and temsirolimus for breast and renal cancers) 7 . Of interest in this regard, PIK3CA-related overgrowth spectrum (PROS) includes a group of congenital disorders that lead to segmental/asymmetric overgrowth of various body parts (brain, limbs, trunk, and face) and anomalies in blood and lymphatic elements as well as in other tissues due to PIK3CA mutations 8 . PROS are caused by heterozygous mosaic postzygotic (or rarely, constitutional) activating pathogenic variants in PIK3CA . Surprisingly, PROS do not predispose to cancers typically related to PIK3CA pathway alterations 9 . In 2022, the FDA approved the PIK3CA inhibitor alpelisib (as mentioned, also approved for breast cancers bearing PIK3CA mutations), for children with PROS, as it combats some of the growth anomalies typical of the disorder 10 . There are many other similar examples of aberrations typical of cancer that can cause non-malignant conditions. For instance, TP53 mutations are detected in ~40% of metastatic cancers and are perhaps the most common tumor suppressor anomaly in cancer 11 . Of interest, TP53 mutations contribute to the overexpression of interleukin-6 (IL-6) in renal cell carcinoma via activation of downstream signaling pathways 12 . Increased IL-6 signaling can also promote chronic inflammatory and autoimmune disease 13 , 14 . Moreover, the synovium from patients with rheumatoid arthritis may bear pathogenic TP53 mutations (without any cancer predisposition) 15 , which may explain why IL-6 levels are elevated in this non-malignant inflammatory disorder and may also account for the effectiveness of the FDA-approved anti-IL6 therapy (tocilizumab; IL-6 receptor antibody) for rheumatoid arthritis 16 . Given that malignant and non-malignant conditions can have similar pathways and genomic drivers, several groups have begun to expand investigation into repurposing drugs from non-malignant to malignant conditions and vice versa 17 - 20 . Herein, we review the spectrum of gene and pathway alterations found in cancer that can also affect non-malignant disease and the agents that have been repositioned or that could be repositioned based on their mechanism of action ( Tables 1 , 2 and 3 ) 4 , 12 , 13 , 17 , 18 , 21 - 71 . Drugs used for adult malignancy/oncology indications were identified on the Chemocare database ( https://chemocare.com/ ) and the National Comprehensive Cancer Network (NCCN) Drugs & Biologics Compendium ( https://www.nccn.org/professionals/drug_compendium/content/ ). The United States Food and Drug Administration (FDA) approved indications for each drug were checked using UpToDate ( http://www.uptodate.com ) and Micromedex ( https://www.micromedexsolutions.com ) and confirmed with drug product labeling from the FDALabel online database ( https://nctr-crs.fda.gov/fdalabel/ui/search ). Therapies that had non-malignant FDA approved indications in addition to malignant indications were included ( Tables 2 and 3 and Table S1 ). The data curated demonstrate that a wide variety of non-malignant disorders, ranging from congenital/hereditary conditions with and without predisposition to cancer, to acquired illnesses such as inflammatory and autoimmune conditions, are caused by alterations in pathways as well as identical alterations in specific genes that also drive cancer, and hence suggest the possibility that targeted drugs developed for cancer may be more widely repurposed for treating non-malignant illnesses ( Figure 1 ).