RNA profiling and immunohistochemistry analyses of circRNAs in imatinib-resistant gastrointestinal stromal tumors

Despite the benefits of imatinib for treating gastrointestinal stromal tumors (GIST), the prognosis for high risk GIST and imatinib-resistant (IR) GIST remains poor. The mechanisms of imatinib resistance have not yet been fully clarified. The aim of the study was to establish imatinib-resistant cell lines and investigate nilotinib, a second generation tyrosine kinase inhibitor (TKI), in preclinical models of GIST and imatinib-resistant GIST. For a model of imatinib-resistant GIST, we generated resistant cells from GK1C and GK3C cell lines by exposing them to imatinib for 6 months. The parent cell lines GK1C and GK3C showed imatinib sensitivity with IC50 of 4.59±0.97 µM and 11.15±1.48 µM, respectively. The imatinib-resistant cell lines GK1C-IR and GK3C-IR showed imatinib resistance with IC50 values of 11.74±0.17 µM (P<0.001) and 41.37±1.07 µM (P<0.001), respectively. The phosphorylation status of key cell signaling pathways, receptor tyrosine kinase KIT (CD117), platelet-derived growth factor receptor alpha (PDGFRA) and downstream signaling kinases: serine-threonine kinase Akt (AKT) and extracellular signal-regulated kinase 1/2 (ERK1/2) or the non-receptor tyrosine kinase: proto-oncogene tyrosine-protein kinase Src (SRC), was analyzed in established cell lines and ERK1/2 phosphorylation was found to be increased compared to the parental cells. Nilotinib demonstrated significant antitumor efficacy against GIST xenograft lines and imatinib-resistant GIST cell lines. Thus, nilotinib may have clinical potential for patients with GIST or imatinib-resistant GIST.

To the Editor: We read with interest the article entitled “A Missense Mutation in KIT Kinase Domain 1 Correlates with Imatinib Resistance in Gastrointestinal Stromal Tumors” by Chen et al. because we have recently found the same mutation in a one of our surgically treated gastrointestinal stromal tumor patients. This patient had a metastatic gastrointestinal stromal tumor carrying a KIT exon 11 mutation (V559A) and underwent surgery because of clinical and radiological disease progression after imatinib treatment. Molecular analyses revealed an exon 11–activating mutation in all but one of the tumoral metastatic nodules analyzed, whose features were consistent with active proliferation and carried the adjunctive exon 13 mutation (T1982C) responsible for Val654Ala substitution. Biochemical analyses revealed a KIT receptor that was equally phosphorylated and expressed in all the specimens regardless of the genotype. The karyotype was normal in all nodules, in line with the findings of Chen et al.However, we would like to comment on the unusually high frequency of the T1982C mutation in their case material (5 of 12 patients), which is strikingly different from the considerably lower percentage of point mutations in gastrointestinal stromal tumors with acquired resistance to imatinib observed by us. Among eight gastrointestinal stromal tumor patients progressing after imatinib treatment characterized by molecular/biochemical and cytogenetic analyses, we identified only two point mutations in two different patients, one responsible for T670I substitution (1) and the other identical to that detected by Chen et al.Although their finding of an association between the mutation (with a normal karyotype) and disease progression under drug treatment is impressive, functional experiments attesting actual KIT/Val654Ala imatinib resistance and its possible kinase-activating effect are strongly recommended to support this assumption. It is only with transfected cells expressing the double KIT mutant (carrying exons 11 and 13 mutations on the same allele) and treated with different drug doses can provide useful information concerning the critical imatinib dose (if any) needed to deactivate/dephosphorylate the mutated receptor. As all their patients were treated with imatinib 400 mg/d, it is not known whether increasing the dose to 800 mg/d may prevent the development of resistant clones. We transfected COS1 cells with KIT/Δ559/T670I and showed that this receptor is resistant to imatinib 15 μmol/L, a clinically unachievable concentration.To acquire further insights into imatinib resistance and test new inhibitory molecules, suitable problem-oriented in vitro experiments coupled with molecular modeling are required.In Response: We appreciate Dr. Tamborini's letter and their thorough and interesting single case report of T670I substitution in one imatinib-resistant gastrointestinal stromal tumor (GIST) patient. We are especially grateful to Dr. Tamborini for sharing with us their recent unpublished finding of the same 1982T→C mutation (Val654Ala) in one of their imatinib-resistant GIST, which further strengthens the in vivo correlation of imatinib resistance and this mutation reported in our study. We will first address the first comment on mutation frequency. Our reported 1982T→C mutation (Val654Ala) represents a very early and a very effective mutation associated with imatinib resistance as shown in our report. One possible reason of our high detection rate of this mutation is the stringent criteria of imatinib resistance and selection of implants (equivalent to in vivo clones) demonstrating true rapid unequivocal progression with estimated doubling time of ≤35 days as shown in Fig. 1 of our report. There is distinctive differences between the truly resistant GIST and the residual GIST. A second reason may be because these five imatinib-resistant GIST implants were surgically resected immediately after detection by computed tomography scans within 3 months of their emergence of imatinib resistance. Fortunately, imatinib resistance is infrequent and the identification of these five cases in our report spanned over 1 year while treating ∼130 GIST patients (106 GIST patients in S00-33 clinical trial and &gt;25 GIST patients treated off protocol). The true rate of Val654Ala mutation in imatinib-resistant GISTs is likely to vary until sufficient cases have been studied to obtain statistically significant estimate. In addition, more than half of our imatinib-resistant GIST patients are not surgical candidate, and biopsy is neither justified nor indicated in the S00-33 protocol, thus preclude molecular studies.Our report presented five cases of convincing temporal and causal association of imatinib resistance and the novel mutation 1982 T→C (Val654Ala) in vivo. Further in vitro studies are in progress. We agree with Dr. Tamborini that transfection experiments using double KIT mutant will provide invaluable in vitro model to study imatinib resistance and develop new drugs.All five GIST patients A to E are refractory to 800 mg/d imatinib. It remains unknown whether starting with 800 mg/d (instead of 400 mg/d) imatinib could have prevented or delayed imatinib resistance. The S00-33 clinical trial (Dr. George Demetri, principal investigator) and translational study results will provide more valuable information and statistics.

Imatinib-resistant gastrointestinal stromal tumors in the era of second- and third-line tyrosine kinase inhibitors: Does surgical resection have a role?

An Intron-Derived Insertion/Truncation Mutation in the BCR-ABL Kinase Domain in Chronic Myeloid Leukemia Patients Undergoing Kinase Inhibitor Therapy

miR-125a-5p regulation increases phosphorylation of FAK that contributes to imatinib resistance in gastrointestinal stromal tumors

Most gastrointestinal stromal tumor (GIST) patients respond to KIT inhibition with imatinib, yet will eventually exhibit resistance. Imatinib-resistance mechanisms are heterogeneous, and little is known about KIT functional roles in imatinib-resistant GIST. Biological consequences of biochemical inhibition of KIT, phosphatidyl-inositol-3-kinase (PI3-K), PLCgamma, MAPK/ERK kinase/mitogen-activated protein kinase (MEK/MAPK), mammalian target of rapamycin (mTOR) and JAK were determined by immunoblotting for protein activation, and by cell proliferation and apoptosis assays in GIST cell lines from imatinib-sensitive GIST (GIST882), imatinib-resistant GISTs (GIST430 and GIST48) and KIT-negative GIST (GIST62). KIT activation was 3- to 6-fold higher in GIST430 and GIST48 than in GIST882, whereas total KIT expression was comparable in these three GIST lines. In addition to the higher set point for KIT activation, GIST430 and GIST48 had intrinsic imatinib resistance. After treatment with 1 muM imatinib, residual KIT activation was 6- and 2.8-fold higher in GIST430 and GIST48, respectively, compared to GIST882. In all GIST lines, cell growth arrest resulted from PI3-K inhibition, and - to a lesser extent - from MEK/MAPK and mTOR inhibition. Inhibition of JAK/STAT or PLCgamma did not affect cell proliferation. Similarly, only PI3-K inhibition resulted in substantial apoptosis in the imatinib-resistant GISTs. We conclude that GIST secondary KIT mutations can be associated with KIT hyperactivation and imatinib resistance. Targeting critical downstream signaling proteins, such as PI3-K, is a promising therapeutic strategy in imatinib-resistant GISTs.

Epithelial-to-Mesenchymal Transition Leads to Docetaxel Resistance in Prostate Cancer and Is Mediated by Reduced Expression of miR-200c and miR-205

Imatinib is a tyrosine kinase inhibitor that is widely used to combat gastrointestinal stromal tumours (GISTs). However, secondary resistance to imatinib is an important challenge in GIST treatment. Recent studies have demonstrated that cancer-derived nanosized exosomes play a key role in intercellular communication, but little is known about the roles of exosomes in imatinib-resistant GISTs. Here, we reveal that exosomes released from imatinib-resistant GISTs transmit drug resistance to imatinib-sensitive tumours. By using iTRAQ technology, we demonstrate that Ras-related protein Rab-35 (Rab35) is upregulated differentially in imatinib-resistant GISTs. Loss of Rab35 decreases exosome secretion, thereby hampering the transmission of imatinib resistance to sensitive tumours. Mechanistically, we showed that the ubiquitin‒proteasome system is involved in elevated Rab35 expression and that ubiquitin-specific protease 32 (USP32), a deubiquitylating enzyme, is bound to Rab35. Further experiments demonstrate that this protease protects Rab35 from proteasomal degradation by reducing Lys48 (K48)-ubiquitination. Additionally, we found that the transcription factor ETV1, which is a lineage survival factor in GISTs, promotes USP32 expression. Collectively, our results reveal that exosomes transmit imatinib resistance in GISTs and that deubiquitylation plays a key role in regulating the transmission process. The USP32-Rab35 axis provides a potential target for interventions to reduce the occurrence of imatinib resistance in GISTs.

Reverse-phase protein arrays (RPPAs) have become an important tool for the sensitive and high-throughput detection of proteins from minute amounts of lysates from cell lines and cryopreserved tissue. The current standard method for tissue preservation in almost all hospitals worldwide is formalin fixation and paraffin embedding, and it would be highly desirable if RPPA could also be applied to formalin-fixed and paraffin embedded (FFPE) tissue. We investigated whether the analysis of FFPE tissue lysates with RPPA would result in biologically meaningful data in two independent studies. In the first study on breast cancer samples, we assessed whether a human epidermal growth factor receptor (HER) 2 score based on immunohistochemistry (IHC) could be reproduced with RPPA. The results showed very good concordance between the IHC and RPPA classifications of HER2 expression. In the second study, we profiled FFPE tumor specimens from patients with adenocarcinoma and squamous cell carcinoma in order to find new markers for differentiating these two subtypes of non-small cell lung cancer. p21-activated kinase 2 could be identified as a new differentiation marker for squamous cell carcinoma. Overall, the results demonstrate the technical feasibility and the merits of RPPA for protein expression profiling in FFPE tissue lysates.

Mechanisms of resistance to imatinib mesylate in gastrointestinal stromal tumors and activity of the PKC412 inhibitor against imatinib-resistant mutants

The standard procedure for processing clinical pathology specimens for diagnostic purposes is by formalin-fixation and paraffin-embedding (FFPE). This method is used world-wide for the vast majority of routine histopathology. Nucleic acids in FFPE specimens are fragmented and crosslinked, changes which interfere with their use in many standard array and other molecular analyses. Massively parallel (or deep or next generation) sequencing (MPS) such as that performed by Illumina's Genome Analyzer and HiSeq and Applied Biosystems SOLiD instruments provide sequence data on millions of short fragments of genomic material in a single assay. DNA used for these assays (including cDNA from gene expression preparations) is prepared for analysis by fragmenting it into short segments. While the short size of fragments required, typically less than 500 base pairs, made it likely that MPS could be applied to DNA and RNA extracted from FFPE tissues, there is little data confirming this or directly comparing results from FFPE and cryopreserved (cryo) specimens. For DNA, issues include whether target enrichment procedures can be applied successfully and whether the use of the FFPE samples would increase variability in target coverage. A special concern for RNA analysis is whether variability in the length of transcripts recovered after FFPE processing would bias calling of differentially expressed genes. We used MPS to analyze specimens processed by splitting the sample and processing mirror images of it immediately by cryo and FFPE. Appropriate kits from Qiagen and Roche were used to extract DNA and RNA from sections of cryo and FFPE tissues. Sequencing libraries were produced using procedures from Illumina and Agilent. Sets of six DNA libraries were constructed for each sample: cDNA from RNA (mRNA), genomic DNA (gDNA), and targeted DNA (tDNA) for matching cryo and FFPE specimens. To determine the nature of the length and termination characteristics of the FFPE RNA, RNA assays used standard mRNA-SEQ and were analyzed with Illumina Genome Analyzer (&amp;gt;36 cycle reads). Sequences of FFPE melanoma and bladder cancer from the RNA-SEQ assay generally started from the 3’ side of the exon and extend about 150-250 bp. The frequency of sequences varied widely among exons, and appears to be informative of RNA levels. Analysis of mirrored images of specimens from patients revealed successful analysis of the FFPE tissues, with comparable numbers of million reads for the mRNA-SEQ libraries (1.01 for cryo vs 0.95 FFPE), gDNA (2.38 cryo vs 2.02 FFPE), and tDNA (0.46 million reads for cryo vs 1.39 million for FFPE). Graphical and statistical comparisons will be presented for these and other samples. Our findings confirm the suitability of FFPE specimens for several key MPS analyses, and indicate that the vast stores of archived specimens will be suitable for intensive genomic analyses. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3938. doi:10.1158/1538-7445.AM2011-3938

Formalin fixation and paraffin embedding (FFPE) is the standard practice to process surgical specimens in preparation for pathological evaluation. The FFPE procedure is known to introduce adverse effects on RNA quality and the ensuing RNA based expression analysis. However, the overall impact of the FFPE procedure alone on the reliability and accuracy of expression data, without influences from additional compromising factors introduced during long-term storage, has not been vigorously assessed. The quality of RNA extracted from recently processed FFPE prostate tissues was examined. FFPE and frozen blocks were prepared from matched surgical specimens and processed in parallel for RNA extraction, two rounds of linear RNA amplification, and genome-wide expression analysis. Expression ratios of prostate cancer versus benign prostatic hyperplasia (BPH) were compared between data derived from the paired FFPE tissues and frozen tissues. RNA extracted from recently processed FFPE prostate tissues was of high quality and suitable for RNA expression analysis. An unbiased analysis of expression data revealed nearly 80% concordance between FFPE tissues and frozen tissues, in genome-wide gene expression differences of cancer versus BPH, and across a wide spectrum of fold expression differences. FFPE procedure itself does not lead to adverse effects on genome-wide expression analysis. Prospective molecular archiving or modified long-term storage conditions should be implemented to expand the utility of FFPE tissues without compromising standard surgical pathology routines.

Despite the success of imatinib in advanced gastrointestinal stromal tumor (GIST) patients, 50% of the patients experience resistance within two years of treatment underscoring the need to get better insight into the mechanisms conferring imatinib resistance. Here the microRNA and mRNA expression profiles in primary (imatinib-naïve) and imatinib-resistant GIST were examined. Fifty-three GIST samples harboring primary KIT mutations (exon 9; n = 11/exon 11; n = 41/exon 17; n = 1) and comprising imatinib-naïve (IM-n) (n = 33) and imatinib-resistant (IM-r) (n = 20) tumors, were analyzed. The microRNA expression profiles were determined and from a subset (IM-n, n = 14; IM-r, n = 15) the mRNA expression profile was established. Ingenuity pathway analyses were used to unravel biochemical pathways and gene networks in IM-r GIST. Thirty-five differentially expressed miRNAs between IM-n and IM-r GIST samples were identified. Additionally, miRNAs distinguished IM-r samples with and without secondary KIT mutations. Furthermore 352 aberrantly expressed genes were found in IM-r samples. Pathway and network analyses revealed an association of differentially expressed genes with cell cycle progression and cellular proliferation, thereby implicating genes and pathways involved in imatinib resistance in GIST. Differentially expressed miRNAs and mRNAs between IM-n and IM-r GIST were identified. Bioinformatic analyses provided insight into the genes and biochemical pathways involved in imatinib-resistance and highlighted key genes that may be putative treatment targets.

3087 Background: Although IM induces durable clinical benefit in pts with metastatic GIST, resistance to this single agent may emerge over time and new approaches are needed. AMN107 is a new agent rationally designed to inhibit in a structurally different manner the tyrosine kinase activities of KIT, PDGFR, and Bcr-Abl. Since AMN107 has demonstrated the ability to inhibit the proliferation of both IM-sensitive and IM-R GIST cells in vitro, a clinical trial was begun to test this agent with rapid assessment of impact using FDG-PET. Methods: Pts with IM-R GIST were treated with AMN107 alone (400 mg p.o. bid), or a combination of IM (400 mg p.o. bid) and inter-cohort dose escalations of AMN107 (200 mg qd, 400 mg qd, and 400 mg p.o. bid). FDG-PET and CT were performed at baseline and after 1 and 4 wks on therapy. Maximum standardized uptake values (SUVmax) were measured in 11 pts in up to 5 lesions with the greatest FDG uptake/pt (n=46 lesions). For each patient, the summation SUVmax (sSUVmax) of all lesions was calculated at each time point. Percentage change in SUVmax and sSUVmax was calculated at wks 1 and 4 compared to baseline. Metabolic response was assessed using EORTC thresholds for % SUVmax change (PR≤ -25% &lt;SD&lt; +25%≤ PD). The longest diameters of all lesions analyzed by PET were measured on corresponding CT images, summed at each time point, and the % change was calculated at wks 1 and 4 compared to baseline. CT response was assessed using conventional RECIST thresholds. Results: At 1 wk, PET imaging of 11 evaluable pts documented metabolic PR in 3/11, SD in 7/11, and PD in 1/11. After 4 wks on therapy, 2 pts whose PET was stable after 1 wk had converted to metabolic PR, and all other imaging results were unchanged. The CT response for all 11 pts was SD after 1 and 4 wks on therapy. Conclusions: These findings suggest that a metabolic response to AMN107, or the combination of AMN107 plus IM, is seen with FDG-PET while CT anatomic response remains SD at 1 and 4 wks. [Table: see text] [Table: see text]

Imatinib resistance is a major obstacle to the successful treatment of gastrointestinal stromal tumors (GIST). Long non-coding RNAs (LncRNAs) have been identified as important regulatory factors in chemotherapy resistance. This study aimed to identify key lncRNAs involved in imatinib resistance of GISTs. First, MSC-AS1 was found to be upregulated in imatinib-resistant GIST tissues and imatinib-resistant GIST cells. Cellular experiments demonstrated that MSC-AS1 overexpression decreased imatinib sensitivity of GIST cells, evidenced by increased cell survival, colony formation, migration, and invasion. Moreover, suppression of MSC-AS1 improved the imatinib resistance of imatinib-resistant GIST cells. Furthermore, MSC-AS1 upregulated the expression of FNDC1 and Anillin via sponging miR-200b-3p, activated the phosphatidylinositol-3-kinase-AKT signaling pathway, and thereby driving imatinib resistance in vitro and in vivo. Overall, this study elucidates the crucial role and mechanism of MSC-AS1 in the imatinib resistance of GIST, providing the potential therapeutic strategy for overcoming the imatinib resistance of GIST.