Molecular Characterization of Acquired Resistance to KRASG12C-EGFR Inhibition in Colorectal Cancer

Abstract

With the combination of KRASG12C and EGFR inhibitors, KRAS is becoming a druggable target in colorectal cancer. However, secondary resistance limits its efficacy. Using cell lines, patient-derived xenografts, and patient samples, we detected a heterogeneous pattern of putative resistance alterations expected primarily to prevent inhibition of ERK signaling by drugs at progression. Serial analysis of patient blood samples on treatment demonstrates that most of these alterations are detected at a low frequency except for KRASG12C amplification, a recurrent resistance mechanism that rises in step with clinical progression. Upon drug withdrawal, resistant cells with KRASG12C amplification undergo oncogene-induced senescence, and progressing patients experience a rapid fall in levels of this alteration in circulating DNA. In this new state, drug resumption is ineffective as mTOR signaling is elevated. However, our work exposes a potential therapeutic vulnerability, whereby therapies that target the senescence response may overcome acquired resistance.

Significance: Clinical resistance to KRASG12C-EGFR inhibition primarily prevents suppression of ERK signaling. Most resistance mechanisms are subclonal, whereas KRASG12C amplification rises over time to drive a higher portion of resistance. This recurrent resistance mechanism leads to oncogene-induced senescence upon drug withdrawal and creates a potential vulnerability to senolytic approaches. This article is highlighted in the In This Issue feature, p. 1.

Figure 1.

Mechanisms of resistance to combined KRAS^G12C and EGFR inhibition in colorectal cancer. A, Graph showing cell viability of parental and resistant (-R) C106 and RW7213 cells. Statistical analyses and P values represent Mann–Whitney test (t test); ****, P d 0.0001. Cmab, cetuximab. B, Heat map of KRAS^G12C and NRAS^G12D alleles detected by single-cell sequencing of C106 resistant subline. CNA, copy-number alteration; VAF, variant allelic frequency; het, heterozygous mutant; NA, not applicable; wt, wild-type. C, FISH staining for the KRAS gene in RW7213 parental cells and resistant subline. Manual review of parental RW7213 cells indicated no amplification [mean KRAS (red)/Cen12 (green) ratio of 1.1; 50 cells counted] in approximately 90% of the hybridized area and approximately 10% hybridized area with increased KRAS copies (mean red/green ratio of 3.5; 50 cells counted). Mean red/green ratio in the resistant subline, based on manual counting of 20 cells, was 6.4 with >20 KRAS (red) signals in all cells. Scale bars, 5 µm. D, Nonsynonymous somatic mutations identified by MSK-IMPACT in the CLR-113 original and resistant PDXs. Mutation types (left) and CCF of mutations identified (right) are color coded according to the legend. SNV, single-nucleotide variant.E, CNAs of the CLR-113 original and resistant PDXs (top). Copy-number log₂ ratios are shown on the y-axis according to the chromosomes on the x-axis. The arrow shows KRAS amplification. F, Plot showing duration of response to KRAS^G12C inhibitor (adagrasib/sotorasib) plus EGFR inhibitor (cetuximab/panitumumab) by patient ID number. Best response by RECIST is noted at the end of each bar, and partial responses are shaded green and stable disease shaded orange. G, Oncoprint of emergent alterations detected in ctDNA of patients with colorectal cancer at the time of radiographic or clinical progression through combined KRAS^G12C and EGFR inhibition. Patient 12 had both ctDNA and tumor tissue analyzed at progression, and emergent alterations identified only in tissue are marked with an asterisk. AMP, amplification.

Figure 2.

Longitudinal analysis of ctDNA in colorectal cancer patients on KRAS^G12C–EGFR inhibition. A–C, Patients with colorectal cancer treated with combined KRAS^G12C and EGFR inhibitors: circles indicate emergent alterations on treatment; bars indicate emergent copy-number changes; and tumor biomarker (CEA) is indicated with a diamond. In all graphs, KRAS^G12C is marked with solid black square and TP53 alterations are marked with a red square to track these truncal alterations. AMP, amplification; GCN, gene copy number.

Figure 3.

Drug withdrawal drives the senescent phenotype in a resistant colorectal cancer cell line with acquired KRAS^G12C amplification. A, Western blot analyses of the effects on MAPK and mTOR pathway regulation in RW7213 parental cells and in RW7213 resistant cells (RW7213-R) with and without cetuximab–sotorasib combination; vinculin is included as a loading control. B, Microscopy images of RW7213-R with and without cetuximab–sotorasib combination: 10× magnification; scale bars, 100 µm. In the bottom left corner, the black square represents the area magnified in the upper-right corner inset. The red arrows indicate senescent cells. C, Ki-67 and β-galactosidase (β-Gal) staining by immunofluorescence (time point 4 days): 10× magnification; scale bars, 100 µm. Quantification represents the percentage of β-Gal– and Ki-67–positive cells per total number of cells; ± symbol indicates variation between pictures. Ten independent pictures have been quantified per condition. D, Western blot analyses of p16, p21, cleaved PARP, and uPAR expression upon drug withdrawal; vinculin is included as a loading control.E, SASP cytokine array time-course experiment. Data shown represent duplicates. Statistical analyses and P values represent two-way ANOVA with Dunnett multiple comparisons test. ns (not significant) = P > 0.05; *, P d 0.05; **, P d 0.01; ***, P d 0.001; ****, P d 0.0001.F, FISH staining for KRAS (scale bar, 5 µm) and IHC for p-ERK, p-S6 (S235), and p16 in tissue samples collected from patient 12, consisting of pretreatment liver metastasis biopsy (pretreatment 2021) and progression liver metastasis biopsy collected 8 days after stopping KRAS^G12C and EGFR inhibitors (resistance 2022). Mean KRAS (red)/Cen12 (green) ratio, based on manual counting of 50 cells from each time point, was 1.8 for the pretreatment specimen and 13.2 for the resistance specimen. p-ERK staining was 2+ involving >90% of cells pretreatment and 3+ involving >90% of cells at progression; p-S6 staining was absent pretreatment and 2+ involving 70% of cells at progression; and p16 staining was 2+ involving 5% of cells pretreatment and 2+ involving 65% of cells at progression. Magnification of all IHC slides is 20×; scale bars, 100 µm.

Figure 4.

Effect of treatment withdrawal in resistant colorectal cancers with amplified KRAS^G12C. A and B, Longitudinal analysis of ctDNA in colorectal cancer patients who held KRAS and EGFR inhibition for approximately 4 weeks after progression. KRAS^G12C ctDNA VAFs are marked with squares, and KRAS plasma copy numbers are marked with circles. All the other variants are reported in green. AMP, amplification; GCN, gene copy number. C, Western blot analyses of p-ERK, p-MEK, p-S6K, p-S6, and cleaved PARP expression upon drug withdrawal and rechallenge with 50 µg/mL cetuximab and 3 µmol/L sotorasib combination; vinculin is included as a loading control. D, Western blot analyses of p-ERK, p-MEK, p-S6K, p-S6, and cleaved PARP expression upon drug withdrawal and rechallenge with 10 nmol/L trametinib (tram); vinculin is included as a loading control. E, Western blot analyses of p-S6K and p-S6 upon drug withdrawal or in drug-containing medium after treatment with 10 nmol/L AZD8055; vinculin is included as a loading control.F, Short-term proliferation assay of RW7213 resistant cells (RW7213-R) in medium containing cetuximab–sotorasib (black) and in senescent conditions (dark red). Cells were seeded in the absence or presence of drugs for 4 days and then treated for 96 hours with increasing concentrations of AZD8055, and then ATP content was measured using CellTiter-Glo. Data represent the average and standard deviation of 3 biological replicates. ns (not significant) = P > 0.05; ^****, P d 0.0001. G, Short-term proliferation assay of RW7213-R in medium containing cetuximab–sotorasib (black) and in senescent conditions (dark red). Cells were seeded in the absence or presence of drugs for 4 days and then treated for 96 hours with increasing concentration of navitoclax, and then ATP content was measured using CellTiter-Glo. Data represent the average and standard deviation of 3 biological replicates. H, Proposed model: KRAS^G12C mutant signaling is maintained at a similar level in parental cells and in resistant cells in the presence of concomitant EGFR and KRAS^G12C blockade. Upon drug removal, KRAS^G12C-amplified signaling drives oncogene-induced senescence characterized by elevated mTOR activity, creating a new steady state that may be targeted by senolytic treatments.