Is the IDH Mutation a Good Target for Chondrosarcoma Treatment?
Current Molecular Biology Reports volume 6, pages1–9(2020)Cite this article
Is the IDH Mutation a Good Target for Chondrosarcoma Treatment? | SpringerLink
Abstract
Chondrosarcomas are rare cancers of bone that arise from the malignant transformation of cells of chondrocytic lineage. They are known to be resistant to systemic cytotoxic chemotherapy and radiotherapy. The mainstay of management of localised disease is en bloc surgical resection with curative intent. Metastatic chondrosarcoma has a dismal prognosis, and to date, there are no proven effective systemic therapies in the advanced setting. Genomic studies have demonstrated that 50 to 80% of chondrosarcomas harbour a mutation in either the IDH1 or IDH2 gene. IDH inhibitors are currently under investigation in clinical trials, after showing promising results in phase 1 studies in IDH mutated cancers. In chondrosarcoma, IDH mutations represent an attractive target, however, early results with IDH inhibitors in IDH mutated chondrosarcoma are modest and the final results of ongoing trials are eagerly awaited.
Role of IDH Family of Proteins in Homeostasis and Mutations in Cancer
Isocitrate dehydrogenase (IDH) is a family of enzymes which play a crucial role in the Krebs cycle. IDH catalyses the oxidative decarboxylation of isocitrate, producing alpha-ketoglutarate (α-KG); nicotinamide adenine dinucleotide phosphate (NADH), the reduced form NADPH; and carbon dioxide [1, 2]. Alpha-KG epigenetically controls gene expression through α-KG-dependant dioxygenase, by inhibition of 2-oxoglutarate(OG)-dependant chromatin-modifying enzymes, including histone demethylases and methylcytosine dioxygenases of the TET family that regulates the cellular epigenetic status [3]. In humans, IDH exists as 3 isoforms. IDH1 and IDH2 are homodimers that utilise nicotinamide adenine dinucleotide phosphate (NADP+) as a cofactor. IDH3 is a heterotetrameric isozyme and employs adenine dinucleotide (NAD+) as a cofactor, resulting in the final product of NADH [2]. The normal functioning of the Krebs cycle, also known as tricarboxylic acid cycle (TCA), is vital for cellular metabolism, redox balance, epigenetic regulations and DNA repair [4, 5].
Mutations in the IDH1 and IDH2 genes are known to be tumorigenic, however, mutations in the IDH3 gene have not been correlated with cancers in humans [https://doi.org/10.1016/j.ccell.2018.04.011 .” href=”https://link.springer.com/article/10.1007/s40610-020-00126-z?fbclid=IwAR1Zlnbif7Of3st7a61jco3y9OKrdRY6LwvtDV_3eaiJJnGZbaS4Hjt-7tk#ref-CR6” id=”ref-link-section-d1460e470″ style=”box-sizing: inherit; color: rgb(0, 75, 131); text-decoration: underline; text-decoration-skip-ink: auto; background-color: transparent; word-break: break-word; overflow-wrap: break-word;”>6,7,8]. The altered amino acids mostly affected are the R132 in IDH1 and the R172 and R140 in the IDH2 gene. The resultant mutated proteins have neomorphic activity, converting α-KG to D-2-hydroxyglutarate (D-2-HG), an oncometabolite, using NADPH as a cofactor [9]. In the normal metabolic state, cellular 2-HG accumulation is limited by endogenous 2-HG dehydrogenase enzymes, which catalyse the conversion of 2-HG to α-KG. In mutated cells, however, the 2-HG can be as much as 100-fold increased and the accumulation of 2-HG affects CpG island and histone methylation, leading to genome-wide histone and DNA methylation alterations [10].
Cancers Associated with IDH Mutations
IDH1 and IDH2 mutations have been identified in solid tumours including glioma, chondrosarcoma, cholangiocarcinoma and haematological malignancies such as acute myeloid leukaemia (AML).
Glioblastoma
Recurrent IDH1 and IDH2 mutations have been identified in more than 80% of low-grade gliomas and secondary glioblastomas (GBM), whilst less than 10% of primary GBM harbour a mutation in the IDH genes [11, https://doi.org/10.1016/j.molmed.2010.07.002 ” href=”https://link.springer.com/article/10.1007/s40610-020-00126-z#ref-CR12” id=”ref-link-section-d1460e525″ style=”box-sizing: inherit; color: rgb(0, 75, 131); text-decoration: underline; text-decoration-skip-ink: auto; background-color: transparent; word-break: break-word; overflow-wrap: break-word;”>12]. Of note, the presence of an IDH mutation is associated with a better prognosis in IDH mutant tumours [13]. Due to the high incidence of IDH aberrations in secondary GBM, mutational status could act as a biomarker to distinguish between primary and secondary GBM.
Acute Myeloid Leukaemia
Whole-genome and exome sequencing studies have revealed recurrent IDH 1 and 2 mutations in 15–20% of newly diagnosed acute myeloid leukaemia (AML). Production of high levels of 2-HG in IDH1 and IDH2 mutated cases leads to DNA hypermethylation via the TET family of proteins. As a result, expression of the Hox family of homeodomain transcription factors that controls morphogenesis and cellular differentiation is increased in association with the presence of IDH mutations and promotes DNA hypermethylation, aberrant cell proliferation and leukaemic cell differentiation [14,15,16]. In a murine model, the combination of IDH1 mutation and the expression of HoxA9 promoted a myeloproliferative phenotype [https://doi.org/10.1016/j.molmed.2010.07.002 ” href=”https://link.springer.com/article/10.1007/s40610-020-00126-z#ref-CR12” id=”ref-link-section-d1460e574″ style=”box-sizing: inherit; color: rgb(0, 75, 131); text-decoration: underline; text-decoration-skip-ink: auto; background-color: transparent; word-break: break-word; overflow-wrap: break-word;”>12]. In contrast to GBM, the presence of an IDH mutation in AML has been associated with a poorer prognosis [17, 18].
Cholangiocarcinoma
Mutations in the IDH1 and IDH2 genes are found in 15–20% of intrahepatic cholangiocarcinomas [19, 20]. Saha and colleagues demonstrated the mechanism of tumorigenesis in transgenic mice, with IDH mutations resulting in the production of 2-HG and suppression of hepatocyte nuclear factor 4 alpha (HNF-4α), an important regulator of hepatocyte differentiation and dormancy status. Hence, mutant IDH will block liver progenitor cells from differentiating into hepatocytes and will promote abnormal cell proliferation and tumorigenesis [https://doi.org/10.1038/nature13441 ” href=”https://link.springer.com/article/10.1007/s40610-020-00126-z#ref-CR21” id=”ref-link-section-d1460e614″ style=”box-sizing: inherit; color: rgb(0, 75, 131); text-decoration: underline; text-decoration-skip-ink: auto; background-color: transparent; word-break: break-word; overflow-wrap: break-word;”>21]. Moreover, co-existence of both KRAS and IDH1 mutations in the same murine tumour promoted aberrant cell proliferation and metastasis of intrahepatic cholangiocarcinoma cells [https://doi.org/10.1038/nature13441 ” href=”https://link.springer.com/article/10.1007/s40610-020-00126-z#ref-CR21” id=”ref-link-section-d1460e623″ style=”box-sizing: inherit; color: rgb(0, 75, 131); text-decoration: underline; text-decoration-skip-ink: auto; background-color: transparent; word-break: break-word; overflow-wrap: break-word;”>21].
Chondrosarcoma
IDH1 or IDH2 mutations have been reported in 50–80% of chondrosarcomas [22, 23]. This review will focus on IDH mutations in chondrosarcoma and its precursors, as well as the latest research in targeting the IDH mutant protein with therapeutic agents.
Current Therapeutic Approaches in Chondrosarcoma
Chondrosarcoma is a rare malignancy, being the 3rd most common bone cancer after myeloma and osteosarcoma. Chondrosarcoma is characterised by a dense extracellular matrix, poor vascularisation and a low percentage of dividing cells. Due to these properties, the metastatic potential of chondrosarcoma is less than that of other cancers [24]. Complete en bloc surgical resection is the gold standard treatment and should be offered whenever feasible. Generally, chondrosarcoma is chemo- and radiation-resistant, and the treatment of advanced disease poses a huge clinical challenge [25, 26].
Conventional chondrosarcomas account for almost 90% of all chondrosarcomas. The majority (85%) of conventional chondrosarcoma are centrally occurring, arising in the medullary cavity of bones, whilst the remaining 15% are peripheral and grow outwards from the surface of the bone.
Other rare subtypes of chondrosarcoma have been described, accounting for 10–15% of chondrosarcomas [27]. The most common subtypes are dedifferentiated (a high-grade non-cartilaginous sarcoma), clear cell, mesenchymal (formed of a chondro-osseous and a blue round cell component), juxtacortical and secondary chondrosarcoma [28]. Myxoid chondrosarcoma of the bone or skeletal myxoid chondrosarcoma (SMC), a distinct entity to extra-skeletal myxoid chondrosarcoma (EMC), is a newly described subtype and is likely a variant of conventional chondrosarcoma [27]. EMC has a higher potential for metastasis than other chondrosarcoma subtypes [29].
As well as demonstrating distinct histological morphologies, these subtypes have variable underlying genetic alterations. Conventional chondrosarcomas present an upregulated Hedgehog pathway, which plays a role in cartilage tumorigenesis [30, 31]. Proto-oncogene tyrosine-protein kinase Src, phosphoinositide 3-kinase (PI3k)/serine-threonine protein kinase Akt/mammalian target of rapamycin (PI3k–Akt–mTOR) and angiogenesis are other activated pathways in conventional chondrosarcomas [32, 33]. The retinoblastoma (RB) and the p53 pathways are commonly affected in clear cell chondrosarcoma, mesenchymal chondrosarcoma and dedifferentiated chondrosarcoma [34]. In these subtypes, chromosomal aberrations are frequent, including loss of the CDKN2A/p16 locus [34].
There is currently no targeted therapy approved for chondrosarcoma, although encouraging results were seen in the phase 2 study of dasatinib, a small molecule tyrosine kinase inhibitor, with an overall response rate (ORR) of 15% and a 6-month progression-free survival (PFS) rate of 47% in chondrosarcoma patients [35, 36]. Vismodegib (previously known as GDC-0449), targeting the hedgehog pathway, has failed to show clinical benefit in a phase 2 trial conducted by Italiano et al.; however, some activity was noted in patients with progressive grade 1 or 2 chondrosarcoma [37]. IPI-926, an antagonist of the Hedgehog pathway, showed growth inhibition of chondrosarcoma xenografts [38]; however, a phase 2 placebo-controlled double-blind randomised study of IPI-926 in chondrosarcoma patients failed to demonstrate a survival benefit. Only a small subset of patients had dimensional tumour reduction in this trial (NCT01310816) [39].
Pazopanib, a multi-targeted tyrosine kinase inhibitor with activity against vascular endothelial growth factor (VEGF) and platelet-derived growth factor receptor alpha (PDGFRA), showed some benefit in terms of prolonged stable disease for patients with progressive chondrosarcoma and demonstrated objective response in a phase 2 trial enrolling patients with extraskeletal chondrosarcoma (NCT01330966, NCT02066285) [40, 41]. Regorafenib, another multi-kinase inhibitor molecule, has been studied in the setting of chondrosarcoma in the non-comparative phase 2 randomised, placebo-controlled clinical trial REGOBONE. The results have been presented at ESMO 2019 and revealed an improved median progression-free survival of 19.4 weeks in the regorafenib arm compared to 8 weeks in the placebo arm, and this difference was maintained at 24 weeks [42].
Other targets such as anti-death receptor 5 (anti-DR5) failed to show a benefit, and the phase 2 trial of anti-DR5 was terminated early due to lack of efficacy in patients with chondrosarcoma (NCT00543712).
Current research is focusing on other molecular targets, with ongoing clinical trials assessing inhibitors of the PI3k-Akt-mTOR pathway (NCT02008019) and histone deacetylase (HDAC) inhibitors in the management of chondrosarcoma (NCT00112463) [43]. Survivin, the antiapoptotic protein, is an attractive target in chondrosarcoma due to its high expression in high-grade chondrosarcomas, and targeting survivin in chondrosarcoma cell lines has shown promise [44, 45].
Glutaminolysis and NAD synthesis pathways are potentially other vulnerable targets in the context of IDH mutated chondrosarcomas, but their clinical applications remain to be determined [46, 47].
IDH Mutations in Chondrosarcoma—Histopathological and Clinical Characteristics
The first documentation of IDH mutations in a mesenchymal tumour was reported in 2011 by Amary et al. The team analysed 145 samples of cartilaginous tumours and found that 56% had a mutation in either IDH1 or IDH2 [48]. No mutations were found in peripheral chondrosarcoma or in the osteochondroma tissues tested, and both low-grade and high-grade chondrosarcomas exhibited the mutation. One striking difference was the anatomic location, with 90% of tumours of the hand and feet harbouring IHD1/2 mutations, compared to only 53.2% of tumours located in the long bones and 53.1% in the flat bones [48]. Of note, this study determined IDH1 mutational status through immunohistochemistry rather than DNA sequencing which may impact on the reliability of these results.
IDH mutations have also been demonstrated in head and neck chondrosarcomas. In a series of 88 cases, IDH mutation was present in 64.5% of craniofacial chondrosarcoma, but was infrequent in laryngeal and tracheal cartilage chondrosarcoma, being mutated in only 11.8% of these cases [https://doi.org/10.1016/j.humpath.2018.09.015 .” href=”https://link.springer.com/article/10.1007/s40610-020-00126-z#ref-CR49” id=”ref-link-section-d1460e809″ style=”box-sizing: inherit; color: rgb(0, 75, 131); text-decoration: underline; text-decoration-skip-ink: auto; background-color: transparent; word-break: break-word; overflow-wrap: break-word;”>49].
Clear cell chondrosarcoma and mesenchymal chondrosarcoma do not harbour IDH1 or IDH2 mutations, as opposed to dedifferentiated chondrosarcoma which harbours a mutation in one of these genes in 50–76% of cases [34, 50, 51].
Due to the high incidence of IDH1 and IDH2 mutations in dedifferentiated and conventional chondrosarcomas, the presence of an IDH mutation could be useful in distinguishing chondrosarcoma from other bone tumours, such as undifferentiated pleomorphic sarcoma (UPS) of the bone. In a study that looked at UPS arising in bone, none of the samples harboured an IDH1 or IDH2 mutation and this characteristic could have important implications due to the different clinical behaviour of UPS and chondrosarcoma [52]. The management of the UPS of the bone comprises neoadjuvant or adjuvant chemotherapy, offering an improved 5-year survival rate up to 59%, as opposed to the poor response to chemotherapy in chondrosarcomas [53]. Amary et al. have also documented that an IDH mutation present in the primary tumour is maintained in local recurrences and metastases from that specific tumour, suggesting that it may be an early driver event in the disease [54].
Maffucci syndrome is a rare congenital non-hereditary disorder presenting as a combination of enchondromatosis (also known as Ollier disease) and haemangiomatosis. Both Maffucci syndrome and Ollier disease interfere with the normal development of tissues leading to physical deformities, and predisposing those affected to secondary neoplasms, such as chondrosarcomas, gliomas, and pancreatic and ovarian cancers, observed in up to 25% of patients with Ollier disease and Maffucci syndrome [https://doi.org/10.1016/j.jons.2017.08.003 .” href=”https://link.springer.com/article/10.1007/s40610-020-00126-z?fbclid=IwAR1Zlnbif7Of3st7a61jco3y9OKrdRY6LwvtDV_3eaiJJnGZbaS4Hjt-7tk#ref-CR55” id=”ref-link-section-d1460e865″ style=”box-sizing: inherit; color: rgb(0, 75, 131); text-decoration: underline; text-decoration-skip-ink: auto; background-color: transparent; word-break: break-word; overflow-wrap: break-word;”>55,56,57,58]. Apart from secondary neoplasms, malignant transformation of enchondromas to chondrosarcomas occurs in > 30% of these patients [59,60,61].
In another study by Amary and colleagues, about 90% of tumours of patients with Ollier disease and Maffucci syndrome harboured an IDH1 or IDH2 mutation [62]. Similar findings were published by Pansuriya et al., where 81% of patients with Ollier disease and 77% of patients affected by Maffucci syndrome carried a somatic mutation of IDH1 (98%) or IDH2 (2%) [63]. Furthermore, mutation of the IDH1 gene was associated with hypermethylation and downregulation of multiple different genes, with their role in the formation of enchondromas and haemangiomatosis yet to be confirmed [63, 64]. Based on the prevalence of IDH mutations in Ollier disease and Maffucci syndrome, it has been suggested that mutated IDH is an early driver event in these conditions and could be the precursor mutation that predisposes the malignant transformation of enchondromas to chondrosarcoma; however, more research is needed to determine the role of IDH mutations in the malignant transformation of enchondromas.
Long-term outcomes did not differ in a series of patients with chondrosarcomas harbouring IDH mutations when compared to IDH wild-type chondrosarcoma [65]. In contrast, IDH1 and IDH2 mutant chondrosarcoma had worse outcomes relative to IDH wild-type chondrosarcoma in another study reported by Lugowksa et al. [66]. The authors of both studies performed correlation with the tumour grade; however, the histological subtype was not investigated. The differences between the two reports could be explained by the lack of histological subtype analysis, as some chondrosarcoma subtypes have a more aggressive clinical course than others. In another recent study, IDH1 and IDH2 mutations were associated with longer relapse-free (measured as the time from surgery to relapse or metastatic disease) and metastasis-free survival (measured as the time from initial diagnosis to the appearance of metastases) in high-grade chondrosarcomas, excluding de-differentiated chondrosarcoma. However, no significant impact of IDH status on overall survival was noted in this study [67].
Targeting the IDH Pathway—Preclinical Data
As previously discussed, mutations of the IDH1 and IDH2 genes result in DNA methylation via the accumulation of D-2-hydroxyglutarate. The most affected hypermethylated genes in IDH mutant samples are retinoic acid receptor alpha, platelet-derived growth factor subunit A (PDGFA) and B cell lymphoma 6 protein corepressor gene (BCOR), suggesting that IDH mutations are associated with epigenetic dysregulation of genes implicated in stem cell maintenance, regulation and dedifferentiation [68]. Expression of the IDH2 mutant protein in a mesenchymal multipotent cell line resulted in high levels of 2-HG, hindered differentiation into adipocytic and chondrocytic cells and resulted in tumour formation in vivo [68].
Preclinical work in the TS603 glioma cell line, characterised by a heterozygous R132H-IDH1 mutation, showed that a selective R132H-IDH1 inhibitor (AGI-5198) blocked the formation of 2-HG. This process induced demethylation of histone H3K9me3, delayed tumoural growth and promoted differentiation of glioma cells [69]. Interestingly, genome-wide DNA methylation was not affected in this experiment, raising the possibility that IDH1 mutation might promote tumorigenesis through other transcriptional mechanisms, independent of DNA methylation.
Similar positive results were noted in AML cell lines. The small molecule inhibitor AGI-6780 was active against R140Q-IDH2 mutant human AML cells in an experiment conducted by Wang and colleagues [70]. The inhibition of R140Q-IDH2 mutant cell by AGI-6780 promoted differentiation of TF-1 erythroleukaemia and human AML cells in vitro and reversed IDH2 mutation-induced histone and DNA hypermethylation [15, 70].
Li and colleagues used AGI-5981 to treat JJ012 (chondrosarcoma cell line) and HT1080 (fibrosarcoma cell line), two human IDH-mutated cell lines and the chondrocyte cell line C28, which is IDH wild type. AGI-5198 decreased the 2-HG levels and inhibited tumour colony formation and migration, inducing apoptosis in the IDH-mutated chondrosarcoma and fibrosarcoma cells [71]. Another experiment using the IDH mutated chondrosarcoma cell lines JJ012, L835, SW1353 and L2975, treated with AGI-5981, resulted in decreased levels of 2-HG, but the tumorigenic properties were not affected, in contrast with previous preclinical work [72].
Clinical Data with IDH Inhibitors
Ivosidenib and Enasidenib: First-in-Class IDH1 and IDH2 Inhibitors
IDH1 and IDH2 mutated proteins have become an attractive target in the treatment of cancers harbouring these mutations [73]. Ivosidenib (previously known as AG-120), is a first-in-class selective oral IDH1 inhibitor that targets the mutant IDH1 protein and was developed by Agios Pharmaceuticals. It showed clinical activity in a phase 1, open-label, multicentre study with 168 patients with advanced solid tumours (NCT02073994) [74, 75]. Enasidenib, also known as AG-221, another IDH inhibitor developed by Agios Pharmaceuticals, is a first-in-class orally potent selective IDH2 inhibitor, targeting the mutant IDH2 enzymes [76]. Ongoing trials with enasidenib are recruiting patients with IDH2 mutations (NCT03515512, NCT03728335).
Toxicity Profile of IDH Inhibitors
The phase 1 data of ivosidenib showed rapid oral absorption and a long half-life of 40 to 102 h after one single administration of the drug [75]. Ivosidenib had an acceptable tolerability profile, with only few dose reductions or discontinuations due to toxicities noted in the phase 1 trial; the most frequent side effects encountered in the AML population were diarrhoea, leucocytosis, febrile neutropenia, nausea, fatigue, QT prolongation, pyrexia and anaemia [77]. In the cholangiocarcinoma cohort of the same trial, the most frequent side effects of ivosidenib reported were fatigue, nausea, diarrhoea, abdominal pain, decreased appetite and vomiting [78, 79]. The chondrosarcoma patients experienced decreased appetite, long QT, nausea, anaemia and peripheral oedema [80]. In the glioma cohort of the same trial, the most frequent adverse events reported were diarrhoea, headaches, nausea, neutrophil decrease and vomiting [81].
Enasidenib has a slightly different toxicity profile. In the phase 1 trial with enasidenib in AML, the most common adverse events were indirect hyperbilirubinaemia and nausea. Grade 3 and 4 adverse events occurred in 99 patients out of a total of 239 evaluable patients (41%), the most common being indirect hyperbilirubinaemia in 29 patients (12%) and IDH-inhibitor-associated differentiation syndrome (IDH-DS) in 15 patients (6%). Other common toxicities were haematological toxicities and infections, leucocytosis and tumour-lysis syndrome [82].
The Activity of IDH Inhibitors in Clinical Trials
In IDH1 mutated AML, ivosidenib showed very encouraging results. The rate of complete remission or complete remission with partial haematologic recovery was 30.4%, and the overall response rate was 41.6% in patients with IHD1 mutated AML [77]. In July 2018, the FDA granted approval for ivosidenib for treatment of patients with IDH1 mutant relapsed or refractory AML [83].
IDH1 mutant cholangiocarcinoma patients had a 6-month progression-free survival rate of 40.1% and median progression-free survival of 3.8 months in the phase 1 trial of ivosidenib [79]. Median overall survival was 13.8 months; however, the results are not mature, as only 66% of patients were censored at the time of publication [79]. The results of the phase 3 study of ivosidenib versus placebo in IDH1 mutated cholangiocarcinoma have been recently presented at the ESMO 2019 congress (NCT02989857) [84]. This randomised placebo-controlled trial enrolled 185 patients with advanced cholangiocarcinoma and IDH1 mutation. The median PFS was improved in the ivosidenib arm, 2.7 months compared to 1.4 months in the placebo arm. More interestingly, the 6- and 12-month PFS rates were 32.0% and 21.9% with ivosidenib, compared to none of the patients on placebo being progression free at 6 months or more [84].
In the chondrosarcoma cohort of the initial phase 1 trial of ivosidenib (NCT02073994), out of twenty evaluable patients, 55% (11 patients) achieved stable disease as their best response. The progression-free survival rate at 3 months was 58%. Four patients had stable disease for more than 6 months. One patient experienced tumour shrinkage, achieving stable disease as best response and remaining on study for 49 weeks [80].
In the phase 1 trial of enasidenib, complete remission was noted in 19.3% of patients with IDH2 mutated relapsed/refractory AML. The overall response rate for AML was 40.3%, and 10% of patients discontinued enasidenib to proceed with stem cell transplant [82]. Enasidenib has received regular approval from the FDA in August 2017 for the treatment of relapsed or refractory acute myeloid leukaemia with an IDH2 mutation detected [https://www.fda.gov/drugs/resources-information-approved-drugs/fda-granted-regular-approval-enasidenib-treatment-relapsed-or-refractory-aml” href=”https://link.springer.com/article/10.1007/s40610-020-00126-z#ref-CR86” id=”ref-link-section-d1460e1299″ style=”box-sizing: inherit; color: rgb(0, 75, 131); text-decoration: underline; text-decoration-skip-ink: auto; background-color: transparent; word-break: break-word; overflow-wrap: break-word;”>86].
The IDH1 inhibitor, FT-2102, is currently being evaluated in a phase 2 trial, with a Simon 2-stage design (NCT03684811). In the first stage, 8 patients with advanced chondrosarcoma will be treated with single-agent FT-2102. If one or more patients have a partial response, then a further 15 patients will be treated with single-agent FT-2102 in the second stage. If no partial responses are documented in the first stage, then chondrosarcoma patients will be treated with combination FT-2102 and 5-azacitidine.
AG-881, a selective oral potent inhibitor of IDH1 and IDH2 mutated proteins, has shown positive results in glioma patients, with one sustained partial response and one minor response confirmed out of 52 patients with glioma enrolled, and another 69% of patients (n = 36) had stable disease as best response [http://investor.agios.com/news-releases/news-release-details/agios-presents-updated-data-phase-1-dose-escalation-study-ag-881” href=”https://link.springer.com/article/10.1007/s40610-020-00126-z#ref-CR85” id=”ref-link-section-d1460e1312″ style=”box-sizing: inherit; color: rgb(0, 75, 131); text-decoration: underline; text-decoration-skip-ink: auto; background-color: transparent; word-break: break-word; overflow-wrap: break-word;”>85]. The most common side effects encountered were mild to moderate and included raised ALT and AST, headaches and fatigue. However, grade 3 or higher side effects were seen in 19% of patients, with one patient experiencing seizures and ALT and AST elevation [http://investor.agios.com/news-releases/news-release-details/agios-presents-updated-data-phase-1-dose-escalation-study-ag-881” href=”https://link.springer.com/article/10.1007/s40610-020-00126-z#ref-CR85” id=”ref-link-section-d1460e1315″ style=”box-sizing: inherit; color: rgb(0, 75, 131); text-decoration: underline; text-decoration-skip-ink: auto; background-color: transparent; word-break: break-word; overflow-wrap: break-word;”>85]. Another drug, IDH305, a selective IDH1R132 inhibitor, was halted in development, due to liver toxicity encountered in AML patients (NCT02826642).
Conclusions
In conclusion, to date, there are two IDH inhibitors which have received FDA approval for the treatment of AML harbouring IDH1 or IDH2 mutations and their toxicity profile is tolerable. The results of the phase 3 trial of ivosidenib in IDH1 mutated cholangiocarcinoma have been presented recently and showed an improved PFS in patients with cholangiocarcinoma harbouring this mutation [84]. However, acquired resistance to IDH inhibitors has already been documented in two AML patients, where a second site IDH2 resistance mutation was found after an initial response to enasidenib [87]. Acquired drug resistance is likely to be the next challenge that needs to be addressed in the treatment paradigm of IDH mutant AML patients.
Treatment options for patients with advanced chondrosarcoma are limited, with no proven effective systemic therapy. The presence of IDH1 or IDH2 mutations in 50–80% of chondrosarcomas suggests that IDH mutations may be a druggable target in this disease. This is supported by promising preclinical data. However, available published clinical data have to a certain extent been disappointing, with disease stabilisation being the best response by RECIST. Advanced chondrosarcoma can be relatively indolent, and the significance of disease stabilisation is therefore difficult to interpret in single-arm trials. Furthermore, there are a number of ongoing trials of IDH inhibitors which include chondrosarcoma. The results of these trials are eagerly awaited and may show that targeting the IDH mutation is effective in chondrosarcomas.
However, if these trials report a lack of efficacy with monotherapy, then combination therapies could potentially be evaluated in carefully stratified trials. Ivosidenib in combination with nivolumab will be investigated in a phase 2 trial which will enrol patients with gliomas and solid tumours with an IDH1 mutation (NCT04056910). Chondrosarcomas express NY-ESO-1 or LAGE-1s in approximately 36% of cases, and this expression was increased in an experiment where chondrosarcoma cells were treated with 5-aza-2-deoxycitabin (5-Aza-dC) [88]. The potential use of combinations of immune checkpoint inhibitors with 5-aza-dc may prove effective in increasing the immunogenicity of chondrosarcomas. Preclinical results in gliomas with IDH1 mutations revealed the presence of CD4+ antitumor T cell response, but such an immune response has not been validated yet in IDH mutated chondrosarcoma [89, 90].
The combination of metformin, an oral antidiabetic medication, and chloroquine, an oral antimalarial drug, has shown activity against the 2-HG oncometabolite, and an ongoing clinical trial using these two agents is enrolling patients with IDH1 or IDH2 mutated solid tumours (NCT02496741) [91].
IDH and 2-HG as Biomarkers in Chondrosarcoma
IDH status is an emerging biomarker in the context of chondrosarcomas. The presence of an IDH mutation could lead to tailored treatment in chondrosarcomas, and the use of 2-HG could potentially correlate with the tumour burden in solid cancers. Further studies are required to ascertain whether this could be used as a predictive marker of response to therapy [92]. Finally, IDH status is correlated with longer relapse-free and metastasis-free survival in high-grade chondrosarcomas, but the impact on overall survival requires further evaluation.