Signature-driven repurposing of Midostaurin for combination with MEK1/2 and KRASG12C inhibitors in lung cancer

Abstract

Drug combinations are key to circumvent resistance mechanisms compromising response to single anti-cancer targeted therapies. The implementation of combinatorial approaches involving MEK1/2 or KRASG12C inhibitors in the context of KRAS-mutated lung cancers focuses fundamentally on targeting KRAS proximal activators or effectors. However, the antitumor effect is highly determined by compensatory mechanisms arising in defined cell types or tumor subgroups. A potential strategy to find drug combinations targeting a larger fraction of KRAS-mutated lung cancers may capitalize on the common, distal gene expression output elicited by oncogenic KRAS. By integrating a signature-driven drug repurposing approach with a pairwise pharmacological screen, here we show synergistic drug combinations consisting of multi-tyrosine kinase PKC inhibitors together with MEK1/2 or KRASG12C inhibitors. Such combinations elicit a cytotoxic response in both in vitro and in vivo models, which in part involves inhibition of the PKC inhibitor target AURKB. Proteome profiling links dysregulation of MYC expression to the effect of both PKC inhibitor-based drug combinations. Furthermore, MYC overexpression appears as a resistance mechanism to MEK1/2 and KRASG12C inhibitors. Our study provides a rational framework for selecting drugs entering combinatorial strategies and unveils MEK1/2- and KRASG12C-based therapies for lung cancer.

Fig. 1. Synergistic dual combinations for mutant KRAS lung cancer obtained through a drug repurposing-based strategy.

A Experimental workflow employed to identify drug combinations with the highest antitumor effect on mutant (mut) KRAS LUAD. B Repurposing scores of drugs obtained from the Connectivity Map at the Library of Integrated Network-based Cellular Signatures (LINCS) program (c3.lincscloud.org) using the interspecies KRAS signature as input. Targets of predicted drugs are shown in brackets. C Combination index (CI) values corresponding to all concentrations for each drug-pair combination tested in mut KRAS cell lines (H1792 and H2009; n = 9). CI < 0.8, synergism. Trametinib (Tram: MEK1/2i); BIX02189 (BIX: MEK5-Kinase2i); Neratinib (Nera: ERRB2i, EGFRi); Lestaurtinib (Lest: PKCi), Dabrafenib (Dabra: BRAFi, CRAFi); Adavosertib (Adavo: WEE1i), Panobinostat (Pano: HDACi). Data: mean +/− SEM. D, E Percent cell viability of H1792 and H2009 cells treated with different concentrations of Tram and Lest (D) or Tram and Adavo (E), individually or in combination (data from the drug screening). F Principal component analysis (PCA) of H1792 cells treated with DMSO (Ctrl), Tram or Lest. G Unsupervised clustering heatmap of iKRASsig genes’ expression in H1792 cells treated with DMSO (Ctrl), Tram or Lest.

Fig. 2. Trametinib and Lestaurtinib combination is preferentially effective in mutant KRAS compared to wild type KRAS LUAD cells.

A Effect of Trametinib (Tram) and Lestaurtinib (Lest) combination on cell viability of wild type (wt) KRAS (H1437, H2126, HCC78, H1993 and H1650) and mutant (mut) KRAS (H1792, H2009, A549, HCC44, H23 and H358) cells treated for 72 h treatment. Tram: 0.5 μM; Lest: 0.625 μM (data: mean +/− SD; test: one-way ANOVA, Tukey’s adjustment). B Effect of Tram and Lest combination on cell viability of mut KRAS LUAD cells (H1792, H2009 and A549) grown in 3D culture conditions (72 h drug treatment; n: 3 cell lines). Tram: 0.01–0.05 μM; Lest: 0.05–0.1 μM (data: mean +/− SD; test: one-way ANOVA, Tukey’s adjustment). C Representative images of mut KRAS LUAD cell lines in (B) exposed to the different treatments. Scale bar: 200 μm. D Percent fold change growth of H1792-derived tumors of mice (n: 8 tumors per group) treated with indicated drugs (data: mean +/− SEM; test: Kruskal–Wallis, Dunn’s adjustment). E Waterfall plot of tumors in (C) at the last day of experiment. Ctrl: untreated control; Tram: 1 mg/kg; Lest: 30 mg/kg; Combo: drug combination. F, G Western blot of cleaved PARP expression of mut KRAS (H1792, H2009 and A549; E) and wt KRAS (H1568, H1993 and H1437; F) cell lines 24 h after drug treatment (loading control: GAPDH).

Fig. 3. Effect of Trametinib and Midostaurin combination on KRAS-mutated LUAD cells.

A H1792 and H2009 percent cell viability after Trametinib (Tram) and Midostaurin (Mido) treatment for 72 h. B Effect of Tram and Mido combination on cell viability of wild-type (wt: H1437, H2126, H1568, H1993 and H1650) and mutant (mut: H1792, H2009, A549, HCC44, H23, H358 and CP435) KRAS LUAD cells (72-h treatment). Tram: 0.5 μM; Mido: 0.625 μM (data: mean +/− SD; test: one-way ANOVA, Tukey’s adjustment). C Effect of Tram and Mido combination on cell viability of mut KRAS cells (H1792, H2009 and A549) in 3D (72-h treatment; n: 3 independent experiments). Tram: 0.01–0.05 μM; Mido: 0.05 μM (data: mean +/− SD; test: one-way ANOVA, Tukey’s adjustment). D Representative images of H1792, H2009 and A549 3D (72-h treatment). Scale bar: 200 μm. E Effect of Tram and Mido on mut KRAS patient-derived xenograft organoids (PDXOs: TP60, TP69, TP80, TP181 and TP126; 72-h drug treatment). Tram: 0.05 μM; Mido: 0.3 μM (data: mean +/− SD; test: one-way ANOVA, Tukey’s adjustment). F, G Cleaved PARP expression in mut (H1792, H2009 and A549; F) and wt KRAS (H1568, H1993 and H1437; G) cell lines after drug exposure (24-h treatment; loading control: β-TUBULIN). H Western blot of indicated proteins in H1792 and H2009 (48-h treatment; loading control: ACTIN). I Long-term assays of H1792, H2009, and A549 cells (10-day treatment; n: 3 independent experiments; data: mean +/− SD; test: one-way ANOVA, Tukey’s adjustment). Crystal violet-stained images of control, Tram- (5 nM), Mido- (100 nM) and combo-treated cells. J Percent cell viability of Tram-resistant (TR) H1792, H2009 and A549 cells (72-h treatment). K Relative depletion of shRNAs against Midostaurin targets in Trametinib-treated versus doxycycline-treated H23 cells (n: 6 shRNAs/gene). Blue: hit selected by Manchado et al.. Red: hits sensitizing to Trametinib with >30% relative depletion. Orange: hits sensitizing between 20% and 30%. Brown: hits sensitizing between 10% and 20%. L Effects of Tram (10–50 nM) and Barasertib (Bara; 25–1000 nM) combination on cell viability of mut KRAS (H1792, H2009, A549, HCC44, H23, H358) cells (72-h treatment; data: mean +/− SD; test: Kruskal–Wallis, Dunn’s adjustment).

Fig. 4. Synergistic effect of KRASi Sotorasib and Midostaurin combination on KRASG12C LUAD cells.

A Cell viability of KRASG12C LUAD cells (H1792, HCC44, H23, H358 and CP435) treated with Sotorasib (Soto; 20–500 nM), Midostaurin (Mido; 0.3–1.25 μM) or both (72-h treatment; data: mean +/− SD; test: one-way ANOVA, Tukey’s adjustment). B Cell viability of KRASG12C cells (H1792, HCC44 and H358) in 3D (72-h treatment). Soto: 62.5 nM; Mido: 0.3 μM (data: mean +/− SD; test: one-way ANOVA, Tukey’s adjustment). C Representative images of H1792, HCC44 and H358 3D cultures exposed to treatments. Scale bar: 200 μm. D Effects of Soto and Mido combination on cell viability of mut KRAS patient-derived xenograft organoids (PDXOs: TP60, TP69, TP80, TP181, TP126) in 3D (72-h treatment). Soto: 62 nM; Mido: 0.3 μM (data: mean +/− SD; test: one-way ANOVA, Dunnet’s adjustment). E Percent cell viability of KRASG12C cells (H1792, HCC44, H23, H358 and CP435) treated with Soto (S), Mido (M), Tram (T), Afatinib (Afati, A) or combinations for 72 h (data: mean +/− SD; test: oneway ANOVA, Bonferroni’s adjustment). F Percent cell viability of KRASG12C cells (H1792, HCC44, and H358) grown in 3D, after 72 h treatment with Soto (S), Mido (M), Tram (T), Afati (A) or indicated combinations (data: mean +/− SD; test: one-way ANOVA, Bonferroni’s adjustment). G Representative crystal violet staining images of 10-day treatment (control: Ctrl; Soto: 12,5-62,5 nM; Mido: 50–100 nM). H Long-term effect of Soto and Mido combination on cell viability of KRASG12C cells (H1792, HCC44, H23 and H358). 10-day treatment (data: mean +/− SD; test: one-way ANOVA, Tukey’s adjustment). I Cell viability percentage of Soto-resistant (SR) H23 and H358 cells treated with indicated concentrations of Soto and Mido, individually or in combination (average of 3 experiments). J Western blot analysis of indicated proteins in H1792 and HCC44 KRASG12C cells (48-h treatments; loading control: ACTIN; exposure time: low and high). K Effects of Soto and Barasertib (Bara) combination on cell viability of mut KRAS (H1792, HCC44, H23, H358) cells (72-h treatment; Soto: 0.02 − 5 µM; Bara: 5 μM; (data: mean +/− SD; test: one-way ANOVA, Tukey’s adjustment).

Fig. 5. Midostaurin-based drug combinations show antitumor effects on treatment naïve and resistant mut KRAS LUAD.

A, B Percent fold change growth of cell-derived tumors (CDXs) from H1792 or A549 cells treated with indicated drugs (Trametinib: 1 mg/kg; Midostaurin: 25 mg/kg). N = 8 tumors per group in Rag2^-/-; Il2γr^-/- mice (data: mean +/− SEM; test: Kruskal-Wallis, Dunn’s adjustment). C, D Waterfall plots of tumors from (A) and (B) at the last day of experiment. E Percent fold change of Tram-resistant (TR) H1792 CDXs treated with indicated drugs (Trametinib: 1 mg/kg; Midostaurin: 25 mg/kg). N = 8 tumors per group in Rag2^-/-; Il2γr^-/- mice (data: mean +/− SEM; test: Mann–Whitney). F Waterfall plot of tumors from (E) at the last day of experiment. G, H Percent fold change of CDXs from H1792 (G) or H358 (H) treated with indicated drugs. 30 mg/kg (H1792) and 10 mg/kg (H358); Mido: 25 mg/kg. N: 14 (ctrl) and 12 (Soto, Mido and Combo) tumors in Rag2^-/-; Il2γr^-/- mice data: mean +/− SEM; test: one-way ANOVA, Tukey’s adjustment). I, J Waterfall plot of tumors in (G) and (H) at the last day of experiment. K Percent fold change growth of Soto-resistant (SR) H358 CDXs treated with indicated drugs (Soto: 10 mg/kg; Mido: 25 mg/kg). N: 10 tumors per group in Rag2^-/-; Il2γr^-/- mice (data: mean +/− SEM; test: t-test). L Waterfall plot of tumors from (K) at the last day of experiment.

Fig. 6. Midostaurin potentiates the antitumor effect of Sotorasib in a GEM lung cancer model.

A Waterfall plot of lung tumors from the KrasG12C driven model (Kras^FSFG12C; Trp53^FRT/FRT mice) treated with the indicated drugs at day 28. N: number of tumors per group. Soto: 100 mg/kg; Mido: 25 mg/kg (test: Kruskal–Wallis, Dunn’s adjustment -all comparisons-; t-test -Soto vs combo-). B Waterfall plot graph of lung tumors from Kras^FSFG12C; Trp53^FRT/FRT mice at day 42 of treatment. *: units correspond to tumor volume change at day 28. N: number of tumors per group. Soto: 100 mg/kg; Mido: 25 mg/kg (test: Mann–Whitney). C Representative microCT images of lung tumors from Kras^FSFG12C; Trp53^FRT/FRT mice at different days of treatment. D Quantification of CD8 staining in tumors from T1-derived xenografts in F1 C57BL/6 x 129S4/Sv mice (7-day treatment). N: 10 tumor sections (ctrl and Mido); n: 6 tumor sections (Soto and Combo). Soto: 30 mg/Kg; Mido: 25 mg/kg (data: mean +/− SEM; test: one-way ANOVA, Tukey’s adjustment). E Immunohistochemistry images of tumors in (D) stained for CD8. Scale bar: 100 µM.

Fig. 7. MtPKCi-based drug combinations downregulate MYC protein.

A Heatmap of biological pathways enriched by the dysregulated proteins obtained from H1792 cell line 48 h after exposure to Trametinib (Tram), Lestaurtinib (Lest) or both, and generated by METASCAPE. B, C MYC protein expression of H1792 and H2009 (B) or H1792 and HCC44 (C) cell lines treated for 48 h. β-TUBULIN (B) or HSP90 (C) are loading controls. Numbers correspond to relative MYC densitometry quantification. D MYC protein expression of H1792 and HCC44 cell lines treated for 48 h (loading control: ACTIN). E, F MYC and p-ERK1/2 protein expression of H23 cell line at different time points (loading control: HSP90). Numbers correspond to relative MYC densitometry quantification. G, H Kaplan–Meier plot showing overall survival of lung cancer patients from TCGA database as a function of the expression of the genes from the signature downregulated by the Tram and Lest combination (ds) and KRAS mutational status (G) or MYC expression and KRAS mutational status (H). I Gene set enrichment analysis (GSEA) of proteins downregulated by combined Tram and Lest treatment onto a ranked-order list from H1792-TR exposed to Tram compared to those cells treated with the drug combination. J GSEA of proteins upregulated by combined Tram and Lest treatment onto the same ranked-order list from (I). K, L MYC protein expression in Tram-resistant H1792 and H2009 cell lines after 48 h drug treatment (loading control: ACTIN). Numbers correspond to relative MYC densitometry quantification.

Fig. 8. MYC upregulation contributes to MEKi and KRASi resistance.

A Kaplan–Meier plot showing overall survival of lung cancer patients from TCGA database as a function of the expression of genes from the Trametinib-resistant upregulated signature (us) and KRAS mutational status. B Heatmap of biological pathways enriched by the dysregulated proteins obtained from Tram-treated parental and Tram-resistant (TR) H1792 cells 48 h after exposure to Trametinib (Tram) using METASCAPE. C MYC, p-MYC S62 and p-MYC T58 proteins’ expression of Parental (Par) and TR H1792 and H2009 cell lines 48 h after exposure to indicated treatments. Ctrl: Control; Tram: Trametinib (loading control: HSP90). Numbers correspond to relative MYC densitometry quantification. D Relative MYC mRNA expression of Parental (Par) and Trametinib-resistant (TR) H1792 and H2009 cell lines 24 h after Trametinib treatment. Housekeeping gene for qPCR: GAPDH (n: 3 independent experiments; data: mean +/− SD; test: two-way ANOVA, Sidak’s adjustment). E MYC protein expression in LacZ or MYC overexpressed H2009 and H358 cell lines (loading control: HSP90). F, G. Percent cell viability of LacZ or MYC overexpressed H2009 and H358 cells after 10-day treatment with increasing concentrations of Trametinib. The IC50 values for Trametinib are shown for each cell line. P values were obtained from the comparison of IC50 values between control (LacZ) and MYC-overexpressing cells (n = 4 experiments; data: mean +/− SD; test: Mann–Whitney -H2009- and t-test -H358-). H Percent cell viability of LacZ- or MYC-overexpressing H358 cells after 10-day treatment with increasing concentrations of Sotorasib. IC50 values for Sotorasib are shown for each cell line. P values were obtained from the comparison of IC50 values between control (LacZ) and MYC-overexpressing cells (n = 4 experiments; test: t-test). I Percent fold change growth of cell-derived tumors (n = 12) from LacZ- and MYC-overexpressing H358 cells treated with Sotorasib (Soto: 10 mg/Kg). (data: mean +/− SEM; test: t-test).