Rationale for co-targeting IGF-1R and ALK in ALK fusion positive lung cancer

Transcription

1 Supplemental Figures and Legends Rationale for co-targeting IGF-1R and ALK in ALK fusion positive lung cancer Christine M. Lovly 1,2, Nerina T. McDonald 1, Heidi Chen 2, Sandra Ortiz-Cuaran 3, Lukas C. Heukamp 4,5, Yingjun Yan 1, Alexandra Florin 4, Luka Ozretić 4, Diana Lim 6, Lu Wang 6, Zhao Chen 7, Xi Chen 2, Pengcheng Lu 2, Paul K. Paik 8, Ronglai Shen 9, Hailing Jin 1, Reinhard Buettner 4, Sascha Ansén 1, Sven Perner 11, Michael Brockmann 12, Marc Bos 3,1, Jürgen Wolf 1, Masyar Gardizi 1, Gavin M. Wright 13, Benjamin Solomon 14, Prudence A. Russell 15, Toni-Maree Rogers 16, Yoshiyuki Suehara 6, Monica Red-Brewer 1, Rudy Tieu 17, Elisa de Stanchina 17, Qingguo Wang 18, Zhongming Zhao 18, David H. Johnson 19, Leora Horn 1, Kwok-Kin Wong 7, Roman K. Thomas 3,4, Marc Ladanyi 6, William Pao 1 Departments of 1 Medicine, 2 Biostatistics, and 18 Biomedical Informatics Vanderbilt University, Nashville, TN, 3 Department of Translational Genomics, Center of Integrated Oncology Köln Bonn, University Hospital Cologne, Cologne, Germany; Departments of 4 Pathology and 1 Internal Medicine (Department I), Center for Integrated Oncology Köln-Bonn, University Hospital Cologne, Cologne, Germany; 5 New Oncology, Cologne, Germany; Departments of 6 Pathology, 8 Medicine, 9 Epidemiology and Biostatistics and 17 Anti-tumor Assessment Core Facility, Memorial Sloan Kettering Cancer Center, New York, New York; 7 Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA; 11 Department of Prostate Cancer Research, Institute of Pathology, Center of Integrated Oncology Köln Bonn, University Hospital of Bonn, Bonn, Germany; 12 Department of Pathology, Hospital Merheim, Cologne, Germany; 13 University of Melbourne, Department of Surgery, St Vincent's Hospital, Melbourne, Australia; 14 Division of Hematology and Medical Oncology, Peter MacCallum Cancer Center, Melbourne, Australia; 15 Department of Anatomical Pathology, St Vincent's Hospital, Melbourne, Australia; 16 Department of Pathology, Peter MacCallum Cancer Center, Melbourne, Australia; 19 Department of Medicine, UT Southwestern School of Medicine, Dallas, TX; 2 Corresponding author: Vanderbilt University Medical Center and Vanderbilt-Ingram Cancer Center, 222 Pierce Avenue, 777 Preston Research Building, Nashville, TN Phone ; Fax: ; christine.lovly@vanderbilt.edu

2 SUPPLEMENTAL FIGURE LEGENDS Supplementary Figure 1: EGFR inhibition does not sensitize ALK+ lung cancer cells to the effects of ALK inhibitors or IGF-1R inhibitors. (a) H3122 cells were treated with erlotinib or erlotinib + MAb391. Soft agar assays were performed using hextuplicate biological replicates. (b) H3122 cells were treated with erlotinib, OSI-96, or the combination. Cell titer blue assays were performed using hextuplicate biological replicates. (c) STE-1 cells were treated with crizotinib or crizotinib + MAb391. Soft agar assays were performed using hextuplicate biological replicates. (d) H3122 cells were treated with crizotinib and OSI-96 as described in Fig. 2b. Data from this experiment were analyzed as previously described 1. The Loewe index provides a measure of drug interaction. A Loewe index < 1 is indicative of a synergistic drug-drug interaction. The blue line in the figure represents a Loewe index of 1. The dashed red lines represent the upper and lower 95% confidence intervals. (e) SUDHL-1 cells were treated with crizotinib, OSI-96, or the combination. Cell titer blue assays were performed using hextuplicate biological replicates. (f) Quantification of H3122 xenograft tumor volumes after treatment with the indicated inhibitors. *P =.41 by the Wilcoxon rank sum test. (g) H3122 cells were treated crizotinib, erlotinib, or the combination. Cell titer blue assays were performed using hextuplicate biological replicates. (h) H3122 cells were treated crizotinib, lapatinib, or the combination. Cell titer blue assays were performed using hextuplicate biological replicates. (i) Quantification of AKT phosphorylation from the western blot shown in Fig. 2f. Supplementary Figure 2: IGF-1 ligand stimulates phosphorylation of IGF-1R but not ALK. (a) H3122 cells were serum starved overnight, treated with vehicle or OSI-96, and then stimulated with IGF-1 for 1min, 5min, or 1min prior to harvest. Lysates were subjected to immunoblotting with antibodies specific for the indicated proteins. (b) Quantification of AKT phosphorylation from the western blot shown in Fig. 3b.

3 Supplementary Figure 3: IRS-1 knock-down impedes the growth of H2228 ALK+ lung cancer cells. (a) H2228 cells were transfected with the indicated sirnas and treated with 5nM crizotinib for 72h.Triplicate biological replicates for each sample were counted on Coulter Counter. Data are representative of two independent experiments. (b) Western blot confirming IRS-1 knockdown in the experiment shown in Supplementary Fig. 3a. Supplementary Figure 4: Characterization of ALK TKI resistant cells. (a) Results from ALK FISH using the ALK Break Apart FISH probe in H3122 CR cells. (b) H3122 parental and isogenic H3122 CR cells were treated with 5nM crizotinib. Lysates were subjected to immunoblotting with the indicated antibodies. (c) Results from ALK FISH using the Vysis ALK Break Apart FISH probe in H3122 XR cells. (d) Lysates from H3122 parental and isogenic H3122 XR cells subjected to phospho-rtk arrays according to the manufacturer s instructions. (e) H3122 CR cells were treated with increasing amounts of crizotinib, OSI-96, or the combination crizotinib + OSI-96 at the indicated concentrations for 72h. Cell titer blue assays were performed to assess growth inhibition. Each point represents hextuplicate replicates. Data are presented as the percentage of viable cells compared to control. (f) Conditioned supernatant from H3122 parental, H3122 CR, and H3122 XR cells were subjected to ELISA assays for IGF-1 according to the manufacturer s instructions. Data are presented as fold change in IGF-1 level compared to H3122 parental cells. Supplementary Figure 5: Efficacy of IGF-1R TKIs in EGFR mutant lung cancer cell lines. Four different isogenic pairs of EGFR TKI sensitive and EGFR TKI resistant cell lines were used in these experiments, including PC-9 parental (TKI sensitive cells) and PC-9 ERc1 (Erlotinib Resistant clone 1) (a), HCC827 parental (TKI sensitive cells) and HCC827 ER (Erlotinib Resistant) (b), HCC2279 parental (TKI sensitive cells) and HCC2279 ER (Erlotinib Resistant)

4 (c), and HCC46 parental (TKI sensitive cells) and HCC46 ER (Erlotinib Resistant) (d). Please see Supplementary Table 1 for a summary of these cell lines. Cell were treated with increasing amounts of erlotinib alone, OSI-96 alone, or the combination of erlotinib + 1µM OSI- 96 for 72h. Cell Titer Blue Assays were performed as previously described. Each point represents hextuplicate biological replicates. Data are representative of two independent experiments. The x-axis indicates the concentration of erlotinib or OSI-96 when these inhibitors were used as single treatment agents. (e) Lysates from the four different isogenic pairs of EGFR TKI sensitive (S) and TKI resistant (R) cell lines described above were resolved by SDS-PAGE and subjected to immunoblotting with the indicated antibodies. Supplementary Figure 6: Representative pigf-1r py1161 immunohistochemistry and Nanostring data from patients with EGFR mutant lung cancer pre- and post-egfr TKI therapy. (a c) Representative pigf-1r py1161 immunohistochemistry showing low (a), medium (b), and high (c) pigf-1r staining on lung tumor biopsies. Scalebars are shown and represent 5 µm. All images viewed correspond to a magnification of 4x. (d f) RNA was extracted from 11 different matched pairs of snap-frozen tumor biopsy samples from patients with EGFR mutant lung cancer pre- and post-egfr TKI therapy. Panel (d) shows aggregate data. 6 pairs of matched tumor samples were from patients who underwent biopsy prior to and 2 days post initiation of an EGFR TKI (e). 5 pairs of matched tumor samples were from patients who underwent biopsy prior to and at the time of acquired resistance to EGFR TKI (f). P values were determined with the paired T-test. Supplementary Figure 7: Effects of LDK-378 in ALK+/TKI sensitive and ALK+/TKI resistant cell lines. (a) STE-1 cells were treated with increasing amounts of crizotinib, LDK- 378, or TAE-684 at the indicated concentrations for 72h. Cell titer blue assays were performed

5 to assess growth inhibition. Each point represents hextuplicate replicates. Data are presented as the percentage of viable cells compared to control (vehicle only treated) cells and are representative of two independent experiments. (b) H3122 XR cells were treated with increasing amounts of X-376 or LDK-378 at the indicated concentrations for 72h. Cell titer blue assays were performed to assess growth inhibition. Each point represents hextuplicate replicates. Data are presented as the percentage of viable cells compared to control (vehicle only treated) cells and are representative of two independent experiments. (c) H3122 cells were treated with crizotinib or LDK-378 for 72 hours prior to harvest. Cells were stained with propidium iodide (PI) and counted on a FACSCalibur machine. Apoptosis was defined by the percentage of cells harboring a sub-2n DNA content. This experiment was repeated two times. Representative results are shown. Supplementary Figure 8: Characterization of novel STE 1 ALK+ lung cancer cell line. (a) Results from ALK FISH using the Vysis ALK Break Apart FISH probe in STE-1 cells. (b) cdna sequencing of ALK from STE-1 cells reveals the presence of an EML4-ALK E13;A2 (variant 1) fusion.

6 SUPPLEMENTAL TABLE LEGENDS Supplementary Table 1: Isogenic pairs of EGFR TKI sensitive and EGFR TKI resistant cell lines. Summary of the isogenic pairs of EGFR TKI sensitive and EGFR TKI resistant cell lines used in these studies. These cell lines have been previously described 2,3. Supplementary Table 2: Microarray data for crizotinib resistant cell lines. Summary of the top 2 downregulated genes in the H3122 crizotinib resistant (CR) cells versus isogenic H3122 crizotinib sensitive parental cells as assessed by microarray analysis. Each value shown for a specific gene represents a unique probe. Changes in IGFBP3 are noted in bold. Supplementary Table 3: Microarray data for X-376 resistant cell lines. Summary of the top 2 downregulated genes in the H3122 X-376 resistant (XR) cells versus isogenic H3122 crizotinib sensitive parental cells as assessed by microarray analysis. Each value shown for a specific gene represents a unique probe. Changes in IGFBP3 are noted in bold. Supplementary Table 4: Summary of the mechanisms of crizotinib resistance evaluated for in the study cohort. a PCR/Direct sequencing; b Whole exome sequencing; c RNA sequencing; d Whole genome sequencing. Supplementary Table 5: Index patient s clinical genotyping results. Results from SNaPshot clinical genotyping of the index patient's tumor. Prior to enrollment in the Profile 17 study of

7 crizotinib vs. pemetrexed or docetaxel for patients with advanced ALK fusion positive lung cancer, the patient s tumor sample was sent for ALK FISH (as part of the study enrollment) and also for our institutional SNaPshot lung cancer panel 4. ALK FISH testing was positive at the central laboratory (Esoterix). No other mutations were found based on the SNaPshot results.

8 SUPPLEMENTAL REFERENCES 1. Boik, J.C., Newman, R.A. & Boik, R.J. Quantifying synergism/antagonism using nonlinear mixed-effects modeling: a simulation study. Stat Med 27, (28). 2. Chmielecki, J., et al. Optimization of dosing for EGFR-mutant non-small cell lung cancer with evolutionary cancer modeling. Sci Transl Med 3, 9ra59 (211). 3. Ohashi, K., et al. Lung cancers with acquired resistance to EGFR inhibitors occasionally harbor BRAF gene mutations but lack mutations in KRAS, NRAS, or MEK1. Proc Natl Acad Sci U S A 19, E (212). 4. Su, Z., et al. A platform for rapid detection of multiple oncogenic mutations with relevance to targeted therapy in non-small-cell lung cancer. J Mol Diagn 13, (211).

9 Supplementary Fig. 1 (Lovly et al 214) a. 1 9 % growth relative to control b. MAB391 1nM erlotinib 5nM erlotinib 1,nM erlotinib 1nM erlotinib + MAb391 5nM erlotinib + MAb391 1,nM erlotinib + MAB391 % growth relative to control Erlotinib OSI-96 Erlotinib + OSI-96 Erlotinib (µm) OSI-96 (µm)

10 Supplementary Fig. 1 (Lovly et al 214) c. d. % growth relative to control MAb391 1nM crizotinib 1nM crizotinib + MAb391 Loewe Index > 1: Drug combination is antagonistic Loewe Index = 1: Drug combination is additive Loewe Index < 1: Drug combination is synergistic

11 Supplementary Fig. 1 (Lovly et al 214) e. 16 f. 7 % growth relative to control Crizotinib OSI-96 Crizotinib + OSI-96 Tumor volume (mm 3 ) * * Vehicle Crizotinib MAb391 Crizotinib + MAb Crizotinib (μm) OSI-96 (μm) Time (days) g. 12 h % growth relative to control Crizotinib Erlotinib Crizotinib + erlotinib % growth relative to control Crizotinib Lapatinib Crizotinib + lapatinib Crizotinib (µm) Crizotinib (µm) Erlotinib (µm) Lapatinib (µm)

12 Supplementary Fig. 1 (Lovly et al 214) i. Pixel intensity compared to control Control IGF-1 Crizotinib OSI-96 IGF-1 + crizotinib IGF-1 + OSI-96 IGF-1 + crizotinib + OSI-96

13 Supplementary Fig. 2 (Lovly et al 214) a. IGF OSI-96 + pigf-1r Y1131 IGF-1R palk Y164 ALK Actin b Control IGF-1 Crizotinib OSI-96 IGF-1 + crizotinib Pixel intensity compared to control IGF-1 + OSI-96 IGF-1 + crizotinib + OSI-96

14 Supplementary Fig. 3 (Lovly et al 214) a % cell number relative to control NT sirna: DMSO NT sirna: Crizotinib IRS-1 sirna (pool #1): DMSO IRS-1 sirna (pool #1): crizotinib IRS-1 sirna (pool #2): DMSO IRS-1 sirna (pool #2): crizotinib b. NT sirna IRS-1 sirna (pool #1) IRS-1 sirna (pool #2) Crizotinib IRS palk Y164 Actin

15 Supplementary Fig. 4 (Lovly et al 214) a. b. Blot: ALK H3122 Parental H3122 CR Crizotinib + + c. d. IGF-1R H3122 Parental H3122 XR

16 Supplementary Fig. 4 (Lovly et al 214) e. f. % growth relative to control Crizotinib OSI-96 Crizotinib + OSI-96 Crizotinib (μm) OSI-96 (μm) Fold change in IGF H3122 parental H3122 CR H3122 XR

17 Supplementary Fig. 5 (Lovly et al 214) a. b. % growth relative to control , Erlotinib (nm) PC-9 parental: erlotinib PC-9 parental: OSI- 96 PC-9 parental: erlotinib + OSI-96 (1μM) PC-9 ER: erlotinib PC-9 ER: OSI-96 PC-9 ER: erlotinib + OSI-96 (1μM) % growth relative to control , Erlotinib (nm) HCC827 parental: erlotinib HCC827 parental: OSI-96 HCC827 parental: erlotinib + OSI-96 (1μM) HCC827 ER: erlotinib HCC827 ER: OSI-96 HCC827 ER: erlotinib + OSI-96 (1μM) c. d. % growth relative to control , Erlotinib (nm) HCC2279 parental: erlotinib HCC2279 parental: OSI-96 HCC2279 parental: erlotinib + OSI-96 (1μM) HCC2279 ER: erlotinib HCC2279 ER: OSI-96 HCC2279 ER: erlotinib + OSI-96 (1μM) % growth relative to control , Erlotinib (nm) HCC46 parental: erlotinib HCC46 parental: OSI-96 HCC46 parental: erlotinib + OSI-96 (1μM) HCC46 ER: erlotinib HCC46 ER: OSI- 96 HCC46 ER: erlotinib + OSI-96 (1μM)

18 Supplementary Fig. 5 (Lovly et al 214) e. HCC- 46 HCC HCC- 827 PC-9 S R S R S R S R IGF-1R pigf-1r Y1131 IRS-1 EGFR Actin

19 5µM Supplementary Fig. 6 (Lovly et al 214) a. b. c.

20 Supplementary Fig. 6 (Lovly et al 214) d. e. Rebiopsy 2 days f. Aggregate data (n=11) post initiation of EGFR TKI therapy (n=6) P =.56 P =.51 Rebiopsy at the time of acquired resistance to EGFR TKI therapy (n=5) P =.84 Log (intensity) Log (intensity) Log (intensity) Pre-EGFR TKI Post-EGFR TKI Pre-EGFR TKI Post-EGFR TKI Pre-EGFR TKI Post-EGFR TKI

21 Supplementary Fig. 7 (Lovly et al 214) a. b. % growth relative to control c Crizotinib 8 7 LDK TAE , Drug concentration(nm) % growth relative to control , Drug Concentration (nm) X-376 LDK Fold change in sub-2n DNA content compared to control Crizotinib LDK-378

22 Supplementary Fig. 8 (Lovly et al 214) a. b. ALK EML4

23 Variant KRAS c. 34G>A (G12S) KRAS c. 34G>C (G12R) KRAS c. 34G>T (G12C) KRAS c. 35G>A (G12D) KRAS c. 35G>C (G12A) KRAS c. 35G>T (G12V) KRAS c. 37G>T (G13C) KRAS c. 37G>A (G13S) KRAS c. 37G>C (G13R) KRAS c. 38G>A (G13D) KRAS c. 38G>C (G13A) KRAS c. 181C>A (Q61K) KRAS c. 182A>G (Q61R) KRAS c. 182A>T (Q61L) KRAS c. 183A>T (Q61H) KRAS c. 183A>C (Q61H) BRAF c.1397 G>T (G466V) BRAF c.146 G>C (G469A) BRAF c.1789 C>G (L597V) BRAF c.1799 T>A (V6E) NRAS c.182 A>T (Q61L) NRAS c.181 C>A (Q61K) NRAS c.182 A>G (Q61R) PIK3CA c.314 A>G (H147R) PIK3CA c.314 A>T (H147L) PIK3CA c.1624 G>A (E542K) PIK3CA c.1633 G>A (E545K) PIK3CA c.1633 G>A (E545Q) MEK1 c.167 A>C (Q56P) MEK1 c.171 G>T (K57N) MEK1 c.199 G>A (D67N) AKT1 c. 49 G>A (E17K) PTEN c.697 C>T (R233X) EGFR c G>T (G719C) EGFR c G>A (G719S) EGFR c G>C (G719A) EGFR c C>T (T79M) EGFR c T>G (L858R) EGFR c T>A (L861Q) EGFR exon 19 deletion EGFR exon 2 insertion ERBB2 exon 2 insertion Result Supplementary Table 1: Results from SNaPshot clinical genotyping from the Index patient's tumor. Prior to enrollment in the Profile 17 study of crizotinib vs. pemetrexed or docetaxel for patients with advanced ALK fusion positive lung cancer, the patient s tumor sample was sent for ALK FISH (as part of the study enrollment) and also for our institutional SNaPshot lung cancer panel. ALK FISH testing was positive at the central laboratory (Esoterix), but no other mutations were found based on the SNaPshot results shown above.

24 PROBE_ID SYMBOL Log Fold Change Adj. p value (H3122 CR vs. H3122 Parental) ILMN_ SERPINE e-7 ILMN_ SERPINA e-6 ILMN_ MAT2A e-7 ILMN_ PRSS e-7 ILMN_ IGFBP e-7 ILMN_ IGFBP e-5 ILMN_ PHLDA e-6 ILMN_ ODC e-7 ILMN_ STMN e-6 ILMN_ ERRFI e-7 ILMN_ ITGB e-7 ILMN_ ITGB e-8 ILMN_ HMGA e-7 ILMN_ FHL e-6 ILMN_ FHL e-7 ILMN_ CYBA e-6 ILMN_ MUC e-6 ILMN_ DUSP e-6 ILMN_ LOC e-5 ILMN_ TRIB e-5 Supplementary Table 2: Summary of the top 2 downregulated genes in the H3122 crizotinib resistant (CR) cells versus isogenic H3122 crizotinib sensitive parental cells as assessed by microarray analysis. Each value shown for a specific gene represents a unique probe. Changes in IGFBP3 are noted in bold.

25 PROBE_ID SYMBOL Log Fold Change adj.p.val (H3122 XR vs. H3122 Parental) ILMN_ TFF e-1 ILMN_ TFF e-8 ILMN_ PLUNC e-8 ILMN_17163 ALPL e-8 ILMN_ CYBA e-8 ILMN_ SLPI e-8 ILMN_ HMGA e-9 ILMN_ CLDN e-9 ILMN_ FCGBP e-9 ILMN_ IGFBP e-7 ILMN_ GDF e-8 ILMN_ DUSP e-8 ILMN_ CFB e-8 ILMN_ PHLDA e-8 ILMN_ CLDN e-8 ILMN_ MUC5AC e-1 ILMN_ SERPINA e-7 ILMN_ C2orf e-8 ILMN_ IGFBP e-8 ILMN_ MUC e-7 Supplementary Table 3: Summary of the top 2 downregulated genes in the H3122 X-376 resistant (XR) cells versus isogenic H3122 crizotinib sensitive parental cells as assessed by microarray analysis. Each value shown for a specific gene represents a unique probe. Changes in IGFBP3 are noted in bold.

26 Established Cell line PC-9 HCC827 HCC2279 HCC46 Summary of isogenic pairs of EGFR TKI sensitive and TKI resistant cell lines Primary Resistance Mechanism EGFR + Resistant EGFR derived of acquired IGF-1R TKIs Cell Line mutation against resistance synergistic? Exon 19 deletion Exon 19 deletion Exon 19 deletion Exon 19 deletion PC-9 ERc1 HCC827 ER HCC2279 ER HCC46 ER Erlotinib EGFR T79M No 2 Erlotinib EGFR T79M No 3 Erlotinib EMT No 3 Erlotinib EMT No 3 Reference (cited in supplemental referenes) Supplementary Table 4: Summary of the isogenic pairs of EGFR TKI sensitive and EGFR TKI resistant cell lines used in these studies. These cell lines have been previously described 2,3.

27 Mechanisms of acquired resistance to crizotinib ALK Patient # pigf-1r IRS-1 ALK mutation amplification 1 + N/A WT a Negative a WT a,b Negative b WT c N/A 4 + N/A G122R d Negative d 5 - N/A WT d Negative d Supplementary Table 5: Summary of the mechanisms of crizotinib resistance evaluated for in the study cohort. a PCR/Direct sequencing; b Whole exome sequencing; c RNA sequencing; d Whole genome sequencing.