Co-clinical trials demonstrate superiority of crizotinib to chemotherapy in ALK-rearranged non-small cell lung cancer and predict strategies to overcome resistance

Abstract

Purpose: To extend the results of a phase III trial in patients with non-small cell lung cancer with adenocarcinomas harboring EML4-ALK fusion.

Experimental design: We conducted a co-clinical trial in a mouse model comparing the ALK inhibitor crizotinib to the standard-of-care cytotoxic agents docetaxel or pemetrexed.

Results: Concordant with the clinical outcome in humans, crizotinib produced a substantially higher response rate compared with chemotherapy, associated with significantly longer progression-free survival. Overall survival was also prolonged in crizotinib- compared with chemotherapy-treated mice. Pemetrexed produced superior overall survival compared with docetaxel, suggesting that this agent may be the preferred chemotherapy in the ALK population. In addition, in the EML4-ALK-driven mouse lung adenocarcinoma model, HSP90 inhibition can overcome both primary and acquired crizotinib resistance. Furthermore, HSP90 inhibition, as well as the second-generation ALK inhibitor TAE684, demonstrated activity in newly developed lung adenocarcinoma models driven by crizotinib-insensitive EML4-ALK L1196M or F1174L.

Conclusions: Our findings suggest that crizotinib is superior to standard chemotherapy in ALK inhibitor-naïve disease and support further clinical investigation of HSP90 inhibitors and second-generation ALK inhibitors in tumors with primary or acquired crizotinib resistance.

Figure 1. Crizotinib is more efficacious against EML4-ALK lung cancer than chemotherapy

A, Waterfall plot showing tumor burden changes in EML4-ALK tumor-bearing mice after three weeks of treatment with daily administration of 50 mg/kg crizotinib. MRI scans reflecting tumor volumes before and after treatment were quantified and compared. Each bar represents one mouse. B, Waterfall plots comparing the response of EML4-ALK-driven tumors to 3 weeks of treatment with 100mg/kg crizotinib or docetaxel or pemetrexed. C, Pharmacodynamic studies showing suppression of signaling pathways downstream of ALK, including pSTAT3, pAkt, pERK and pS6. Tumors were harvested within 24 hours after 2 doses of crizotinib at 100 mg/kg and assessed by immunohistochemistry. Quantification represents the average intensity over 10 microscopic fields, as described in the materials and methods. D, Summary of the treatment study in panel B.

Figure 2. Crizotinib produces improved overall survival and progression-free survival in EML4-ALK lung cancer compared with chemotherapy

A, Kaplan-Meier curve showing overall survival of mice treated with crizotinib, docetaxel or pemetrexed. Total survival times are counted from treatment initiation. B, Kaplan-meier curve showing progression-free survival of mice treated with crizotinib, docetaxel or pemetrexed. Disease progression is defined as described in the methods.

Figure 3. The HSP90 inhibitor 17-DMAG is a candidate to overcome primary crizotinib resistance

A, 17-DMAG suppresses tumor metabolism to a greater extent than crizotinib. PET-CT was used to assess changes in ¹⁸FDG uptake after 2 treatments of either 17-DMAG or crizotinib in EML4-ALK tumor bearing mice. SUVmax for each individual tumor nodule before and after treatment was compared to calculate relative changes. Each bar represents a discernible PET positive tumor nodule. B, Statistical summary of results presented in panel A. C, Representative images showing reduction but not complete elimination of ¹⁸FDG uptake after 2 doses crizotinib in an EML4-ALK tumor-bearing mouse. Further treatment with 17-DMAG completely ablates the PET signal.

Figure 4. The HSP90 inhibitor 17-DMAG and the second-generation ALK inhibitor TAE684 overcome acquired crizotinib resistance

A, EML4-ALK L1196M mutant responds to both 17-DMAG and TAE684. Representative MRI images showing tumor regression after 2 weeks treatment with either drug are shown. B, Both 17-DMAG and TAE684 can overcome crizotinib acquired resistance conferred by the secondary kinase domain mutation F1174L. Four mice with lung cancer driven by EML4-ALK F1174L (each indicated by a different color) were treated with 17-DMAG. Tumor burden was monitored by MRI at the indicated time points from time of treatment initiation. Treatment was switched to TAE684 at the time point marked by the red arrow, after development of acquired resistance to 17-DMAG. Tumor shrinkage followed by re-growth was observed at 11, 3 and 5 weeks after TAE684 administration, respectively, in 3 out of the 4 mice. C, ¹⁸FDG PET-CT study shows TAE684 can overcome acquired resistance to 17-DMAG in EML4-ALK lung cancer. Representative images are shown from a mouse treated with 17-DMAG until development of acquired resistance as reflected by recurrent ¹⁸FDG uptake (bottom left panel). Application of TAE684 successfully suppressed the recurrent 18FDG signal (bottom right panel). D, Upper panels, pharmacodynamic studies showing restoration of known signaling pathways downstream of ALK in F1174L mice that have become resistant to both 17-DMAG and TAE684 as single agents, including pSTAT3, pAkt, pEMK, pERK and pS6. Consistent areas within the same tumor nodule are shown. Lower panels. Tumors from untreated tumor-bearing mice sacrificed without any treatment applied were subjected to IHC. IHC quantification represents the average intensity over 10 microscopic fields.