Characterizing Secondary-Site Mutations in Isocitrate-Dehydrogenase-1 (IDH1)

Background

Isocitrate Dehydrogenase 1 (IDH1) mutations are found in 6-10% of acute myeloid leukemia (AML) patients, leading to a gain-of-function that converts α-ketoglutarate (α-KG) to R-2-hydroxyglutarate (R-2HG). This metabolic shift promotes epigenetic changes favouring hypermethylation and inhibition of cellular differentiation. While initial success has been achieved with allosteric inhibitors targeting mutant IDH1 (mIDH1), resistance has become a challenge, prompting further exploration of alternative treatment approaches.

Methods

Our study aims to explore the mechanisms of IDH1 resistant mutations in AML and their impact on disease progression and treatment outcomes. Using HoxA9-immortalized mouse bone marrow cells transduced retrovirally with mIDH1, including clinically relevant second-site mutations (H315D, G289D, S280F, R119P), we conducted in vitro analyses of sensitivity to a repertoire of mIDH1 inhibitors by evaluating intracellular R-2HG levels via mass spectrometry (LC/MS) and proliferation rates. The inhibitors investigated include ivosidenib (AG120), IDH889, IDH305, DS-1001b, GSK864, BAY1436032, olutasidenib (FT-2102), and vorasidenib (AG881), some of which either are in clinical trials or are FDA-approved. Around 1 million mIDH1 expressing cells were transplanted into lethally irradiated syngeneic C57BL/6 mice to assess their leukemogenic potential.

Results

In vitro the proportion of cells in S/G2/M phases of cell cycle were similar for the double mutants compared to the single mutant R132H except for the mutant G289D/R132H with a significantly lower proportion of cycling cells. The doubling time for the G289D/R132H mutant was significantly longer compared to the single mutant and all other double mutants.

In vivo transplantation of these mIDH1 second-site mutants and the R132H mutant into lethally irradiated mice revealed differences in leukemic latency and phenotype. The engraftment of leukemic cells in peripheral blood 4 weeks after transplantation was 77, 75, 60, 13, and 27% for mIDH1, H315D/R132H, G289D/R132H, S280F/R132H, and R119P/R132H, respectively. The median survival was 29, 29, 43, 68, and 61 days for these groups, respectively, demonstrating the same leukemogenic potential of the H315D/R132H variant compared to the single mutant, while the other mutants exhibited slower disease kinetics. At sacrifice, mean WBC count was 24, 14, 13, 32, and 18 (x10^{^}3/µl), mean spleen weight was 271, 224, 339, 373, and 488 mg, and mean platelet count was 33, 63, 43, 291, and 79 (x10^{^}3/µl) in mIDH1, H315D/R132H, G289D/R132H, S280F/R132H, and R119P/R132H groups, respectively (n=6). The proportion of bone marrow cells at sacrifice with a monocytic phenotype (CD11b⁺) was 0.65, 0.33, 0.19, 26 and 27% in mIDH1, H315D/R132H, G289D/R132H, S280F/R132H, and R119P/R132H groups, respectively, suggesting that S280F/R132H and R119P/R132H mutants shift the cells to a more differentiated phenotype.

We next evaluated the drug sensitivity pattern of single and double mutant cells to the known IDH1 inhibitors. We provide a matrix of mIDH1 with and without second site mutants for the tested inhibitors regarding 50% inhibitory concentration (IC50) based on proliferation and R-2HG reduction. In S280F/R132H cells, DS-1001b and IDH305 reduced cell proliferation (IC50: 125nM and 81nM, respectively) and showed greater sensitivity to olutasidenib and DS-1001b for R-2HG reduction (IC50: 9nM and 31nM respectively) compared to R132H cells. Similarly, DS-1001b inhibited proliferation in R119P/R132H mutant cells (IC50: 321nM) and reduced R-2HG levels (IC50: 10nM). Notably, R119P/R132H cells were also sensitive to vorasidenib for R-2HG reduction (IC50: 13nM). For H315D/R132H cells strong antiproliferative activity was observed with GSK864 (IC50: 7.4nM) and vorasidenib (IC50: 32nM), while olutasidenib and vorasidenib achieved strong R-2HG inhibition (IC50: <0.001nM and <0.001nM, respectively).

Conclusion

Second-site mutants of mIDH1 exhibit different leukemogenic properties and differentiation potential in vivo, and show distinct patterns of resistance and sensitivity to ivosidenib, olutasidenib and other mIDH1 inhibitors. Our study provides a rationale to choose an IDH1 inhibitor in patients with secondary site IDH1 mutations.