Clinical Trial on IDH1 Inhibitor Shows Success in Decreasing Chondrosarcoma Cells

Treatment with a Small Molecule Mutant IDH1 Inhibitor Suppresses Tumorigenic Activity and Decreases Production of the Oncometabolite 2-Hydroxyglutarate in Human Chondrosarcoma Cells

Jonathan Trent, M.D., Ph.D. is a world renown leader of the Sarcoma, Associate Director, Clinical Research, University of Miami, Sylvester Comprehensive Cancer Center, he currently runs a research lab and is conducting clinical trails on Chondrosarcoma. The Abstract and Discussion of the Clinical Trial on the IDH1 inhibitor AGI-5198 is listed below.

For more information the link to the article is:

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0133813

Abstract

Chondrosarcomas are malignant bone tumors that produce cartilaginous matrix. Mutations in isocitrate dehydrogenase enzymes (IDH1/2) were recently described in several cancers including chondrosarcomas. The IDH1 inhibitor AGI-5198 abrogates the ability of mutant IDH1 to produce the oncometabolite D-2 hydroxyglutarate (D-2HG) in gliomas. We sought to determine if treatment with AGI-5198 would similarly inhibit tumorigenic activity and D-2HG production in IDH1-mutant human chondrosarcoma cells. Two human chondrosarcoma cell lines, JJ012 and HT1080 with endogenous IDH1 mutations and a human chondrocyte cell line C28 with wild type IDH1 were employed in our study. Mutation analysis of IDH was performed by PCR-based DNA sequencing, and D-2HG was detected using tandem mass spectrometry. We confirmed that JJ012 and HT1080 harbor IDH1 R132G and R132C mutation, respectively, while C28 has no mutation. D-2HG was detectable in cell pellets and media of JJ012 and HT1080 cells, as well as plasma and urine from an IDH-mutant chondrosarcoma patient, which decreased after tumor resection. AGI-5198 treatment decreased D-2HG levels in JJ012 and HT1080 cells in a dose-dependent manner, and dramatically inhibited colony formation and migration, interrupted cell cycling, and induced apoptosis. In conclusion, our study demonstrates anti-tumor activity of a mutant IDH1 inhibitor in human chondrosarcoma cell lines, and suggests that D-2HG is a potential biomarker for IDH mutations in chondrosarcoma cells. Thus, clinical trials of mutant IDH inhibitors are warranted for patients with IDH-mutant chondrosarcomas.

Discussion

Chondrosarcoma is the second most common primary malignant tumor of the skeleton, and is particularly problematic since most subtypes are resistant to chemotherapy or radiotherapy. Thus, development of novel treatments are critical to patients with tumors that are not amenable to surgical resection. The recent discovery that many chondrosarcomas have a mutation in IDH enzymes offers hope that targeted IDH inhibitors may provide a novel treatment strategy and improve outcomes for patients.The work of Rohle et al and Fathi et al first demonstrated the relationship between these mutations and tumorigenesis, and established D-2HG as a biomarker for IDH-mutant cancer cell activity [24, 38]. Using AGI-5198, Rohle and colleagues demonstrated that mutant IDH1 inhibition blocked production of D-2HG by the mutant enzyme in a dose-dependent manner, and inhibited tumor growth in mouse xenografts and promoted cell differentiation in glioma cells [38]. In our study, we aimed to demonstrate that AGI-5198 would have similar effects on IDH1-mutant human chondrosarcoma cell lines JJ012 and HT1080. We found that AGI-5198 decreased the levels of D-2HG within cells and the conditioned cell medium in a dose-dependent manner in the two chondrosarcoma cell lines. In fact, D-2HG level was undetectable after treatment with 10μM or 20μM AGI-5198. This is consistent with Rohle’s findings which documented that AGI-5198 decreased D-2HG level dose-dependently in R132H-IDH1 mutant TS603 glioma cells. These results provide direct evidence that this compound, identified to inhibit mutant R132H IDH1 enzyme, is also able to inhibit R132G and R132C mutations in IDH1. L-2HG, the stereoisomer of D-2HG, was not detectable in any of our samples, consistent with previous observations that L-2HG is unlikely contributing to tumorigenesis from mutant IDH1. Thus, this supports our hypothesis that D-2HG can serve as a biomarker of IDH-mutant enzyme activity in chondrosarcomas.

In further support of this concept, we observed a decrease in D-2HG levels in post-resection blood and urine samples of a patient with IDH2 R172S mutated dedifferentiated chondrosarcoma. Interestingly, the level did not fall to undetectable which has been demonstrated in IDH-mutant leukemia patients achieving complete remission after chemotherapy [24]. Further investigations will be needed to determine whether this could be a result of prolonged clearance from the patient, limitations in the assay, or whether the patient could still have occult micrometastatic disease. The upcoming clinical trials of mutant IDH1 inhibitor AG-120 in solid tumors which include measurement of D-2HG levels will likely shed light on these important clinical questions.

While this decrease in D-2HG supported that the mutant IDH1 enzyme was being inhibited with AGI-5198 treatment in our chondrosarcoma cell lines, the next question was if we could demonstrate corresponding anti-tumor activity with the compound, similarly to the results seen in gliomas. Interestingly, we did not see short-term inhibition of cell viability or profound early induction of apoptosis with AGI-5198 treatment in chondrosarcoma cells. However, significant inhibition in long-term tumorigenic activity was witnessed based on colony-forming potential. Additionally, the migration of these cells was significantly altered, suggesting that inhibition of mutant IDH1 is clearly disrupting chondrosarcoma cell physiology.

Regarding possible mechanisms to explain our observations, previous studies have reported altered cell differentiation via effects on methylation status after mutant IDH inhibition in cancer cell lines. For example, Wang et al showed that targeted inhibition of mutant IDH2 in leukemia cells reversed the histone and genomic DNA methylation state and induced cellular differentiation [42]. Rohle et al found under conditions of near-complete D-2HG inhibition by AGI-5198, the mutant IDH1 inhibitor induced demethylation of histone H3K9me3 and expression of genes associated with gliogenic differentiation [38]. Based on these findings, we hypothesize that AGI-5198 preferentially affects migration due to the induction of wide scale epigenetic changes, possibly leading to cell differentiation. Given the relatively slow cell cycle of chondrosarcoma cells, and the well-reported phenomenon of the need for numerous cell cycles for epigenetic alterations to show effects, this would explain why we were not seeing profound changes in chondrosarcoma cell growth in the relatively short term assays used in this exploration. The investigation into AGI-5198-mediated changes in gene expression, epigenetics and differentiation of IDH-1 mutant chondrosarcoma cell lines is ongoing in our laboratory.

Most cancer chemotherapeutics have historically relied on induction of apoptosis or interruption of the cell cycle, generally through production of DNA damage that then triggers cell cycle arrest and apoptotic signaling [2]. With the new development of targeted therapeutics like mutant IDH, anti-tumor effects are exerted through alternative pathways, such as disruption of cancer cell metabolism. Thus, this would explain why our results showed only minimal impact on the induction of apoptosis and only slight cell cycle changes with AGI-5198 treatment. A lack of early decrease in cell viability after 72 h of incubation was also reported in early preclinical testing of AG-120 in human leukemia cells (Agios investigator’s brochure). Additionally, despite not producing cell death at lower concentrations, our results confirmed that AGI-5198 significantly decreased the migration distance at concentration much lower concentration of 1 μM in JJ012 and HT1080. This could suggest the drug is particularly potent at inhibition of the metastatic phenotype of chondrosarcoma cells. Our interpretation of these and other’s results is that investigation of pre-clinical effects of mutant IDH inhibitors will likely require alternative assays focusing on longer-term outcomes rather than the traditional rapid response seen in more potent chemotherapeutics that induce apoptosis. Increased tumor latency in xenograft models, and ultimately the response in human clinical trials showing prolonged time to progression may be the more accurate way to assess the therapeutic potential of mutant IDH inhibition and similarly directed novel therapies.

In conclusion, our study investigated in vitro characteristics of IDH-mutated human chondrosarcoma cells and demonstrated anti-tumor activity of a mutant IDH1 inhibitor, AGI-5198. As mentioned before, mutant IDH enzymes fail to convert isocitrate into -KG and instead gain the function to catalyze α-KG into the oncometabolite D-2HG. 2HG, at high levels, inhibits α-KG-dependent dioxygenases including epigenetic regulators, which results in altered epigenetic state of DNA and histone, and promotes tumorigenesis [16, 26, 27, 43]. Therefore, we believe that is why the anti-tumorigenic activity of AGI-5198 was most obvious at high drug concentrations at which the 2HG production was nearly completely blocked. Our data supports that D-2HG is likely to serve as a biomarker for mutant IDH activity in chondrosarcomas, and supports additional investigation of mutant IDH1 inhibition as a treatment strategy in patients with unresectable disease. Future work will be directed at unraveling the complex mechanisms of mutant IDH activity in chondrosarcoma and may improve understanding of how to best utilize therapies that impair cancer cell metabolism, rather than induce rapid apoptosis. Clinical trials investigating mutant IDH inhibitors in solid tumors with IDH mutations have the potential to drastically change standard therapy for chondrosarcomas, and unlock an entirely novel approach to treatment of cancers.