Tyrosine Kinase Inhibitor Treatment Failure Research Articles

Mutation of Epigenetic Regulators at Diagnosis Is an Independent Predictor of Tyrosine Kinase Inhibitor Treatment Failure: A Report from the Residiag Study

Introduction Additional mutations at chronic myeloid leukemia (CML) diagnosis have been shown to variably affect tyrosine kinase inhibitor (TKI) response and were inconstantly detected at loss of response. Contradictory observations may have resulted from difficulties in reliably inferring CML clonal architecture from mutations quantified by NGS, BCR::ABL1 by qRT-PCR, ABL1-TK by RNA-Seq. The RESIDIAG study was designed to identify at diagnosis molecular markers predictive of TKI treatment failure and analyze the clonal dynamics using asymmetric capture sequencing (aCAP-Seq). Methods DNA samples from 60 TKI responders (median follow-up: 7.1 years) were sequenced at diagnosis by aCAP-Seq, allowing quantification and breakpoint sequencing of genomic BCR::ABL1 fusion together with 43 myeloid genes, including ABL1 with a limit of detection of ~0.1% for BCR::ABL1 and 0.5-1% for SNVs and Indels. In parallel, CML patients who experienced treatment failure (ELN 2020) were sequentially analyzed at diagnosis and at failure (n=53), at diagnosis only (n=1), and at failure only (n=3). Small deletions and BCR::ABL1 breakpoints were analyzed to search for microhomology-mediated end-joining (MMEJ) repair signatures. Mutation analysis followed strict medical standards. Results The proportion of patients receiving imatinib at diagnosis wasn't statistically different (p=0.08; Pearson's Chi-square test) between responders (70%) and the failure group (56%). At diagnosis, the number of mutations was higher (p<0.001, t-test with Welch correction) in the failure group (0.9±0.1) as compared with responders (0.18± 0.05). ASXL1, DNMT3A, and TET2 were more frequently mutated at diagnosis in the failure group (38.6%, p<0.001; 10.5%, p=0.01, and 4.3%, p=0.025 respectively; Pearson's Chi²test) as compared with responders (6.7%, 0% and 0% respectively) and the presence of a mutation in one of these genes was highly associated with a reduced failure-free survival (p<0.001, Log-ranked test, Figure 1). An MMEJ signature at BCR::ABL1 breakpoints was more frequent (p=0.014, Pearson's Chi² test) in the failure group (29.1%) than in the responder group (9.1%). and was significantly associated with the ELTS score (p=0.005, Pearson's Chi² test) but not the Sokal score (p=0.017, Pearson's Chi² test). At first failure, 69% of patients had additional mutations in ASXL1 (39%), ABL1-TK (38%), DNMT3A (18%), or TET2 (15%), while other genes were rarely (<5%) mutated. ASXL1 mutations found at diagnosis were still detectable in 18/22 patients at failure and VAF suggested that double mutant BCR::ABL1/ ASXL1 clones were driving TKI failure with or without BCR:: ABL1 tyrosine kinase domain mutation (TKD). ASXL1 mutations disappearing upon treatment were mostly at low frequency <5% at diagnosis and thus possibly not clonally related to CML. Furthermore, a stepwise selection of variables followed by multivariate analysis with the Fine-Gray model identified the age of patients (sdHR [95% C.I.] = 0.99 [0.97-1.00]; p=0.08) and the presence of a mutation in epigenetics regulators at diagnosis as factors affecting the onset of TKD mutations (sdHR = 3.01 [1.76-5.15]; p<0.001). Intriguingly, co-occurring mutations in DNMT3A and TET2 were found in five patients at TKI failure. Their allelic frequency kinetics suggested that these belonged to the same clone driving CML relapse in 4/5 CML patients, while one patient also had a TKD mutation likely involved in CML relapse. Finally, multivariate COX regression analysis identified two independent predictors of TKI failure at diagnosis: a high-risk ELTS score (p=0.001; hazard ratio=3.42 [1.62-7.22] with low-risk as baseline) and a mutation in ASXL1, DNMT3A or TET2 (p=0.002; hazard ratio=2.87 [1.49-5.51]). Conclusion Altogether these results strongly point to the contribution of epigenetic regulator mutations in the emergence of TKI failure in CML and warrant further biological studies to better understand the underlying mechanisms

Efficacy and influencing factors of immunotherapy combined with chemotherapy and bevacizumab in patients with non-small cell lung cancer after epidermal growth factor receptor tyrosine kinase inhibitors treatment failure

Objective: To investigate the efficacy and influencing factors of immunotherapy combined with chemotherapy and bevacizumab in patients with non-small cell lung cancer (NSCLC) who failed epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) treatment. Methods: A retrospective analysis was made on the clinical data of 60 NSCLC patients who were treated with immunotherapy combined with chemotherapy and bevacizumab after EGFR-TKIs treatment failure in the Affiliated Cancer Hospital of Shandong First Medical University from January 2019 to March 2022. Patients were followed up by telephone or outpatient review up to October 1, 2022, with a median follow-up of 8.2 months (95%CI: 7.1-9.3). All 60 patients were followed up. The response evaluation criteria in solid tumors were used to evaluate the short-term efficacy. The adverse reactions of patients were evaluated according to the common terminology criteria for adverse events. The survival curve was drawn by Kaplan-Meier method. Cox proportional hazard regression models were utilized to analyze the influencing factors of progression-free survival (PFS). Results: Among the 60 NSCLC patients, 22 were males. The age ranged from 41 to 75 years, with a median age of 61 years. Eleven patients had partial response, 19 patients had stable disease and 30 patients had progressive disease. The median PFS was 8.2 months (95%CI: 7.2-9.2). The median PFS of patients with low expression of programmed death receptor-ligand 1 (PD-L1) [Tumor cell Proportion Score (TPS)<1%], moderate expression of PD-L1 (1%≤TPS≤49%), and high expression of PD-L1 (TPS≥50%) were 6.4 (95%CI: 4.8-8.0), 8.3 (95%CI: 7.3-9.3) and 10.6 months (95%CI: 7.2-14.1), respectively, and there were statistically significant differences (χ2=13.58, P<0.001). Multivariate Cox proportional risk regression model analysis showed that age>65 years old (HR=4.017, 95%CI: 1.468-10.992, P=0.007) was a risk factor for PFS in NSCLC patients who received immunotherapy combined with chemotherapy and bevacizumab after EGFR-TKIs treatment failure. Moderate expression of PD-L1 (HR=0.360, 95%CI: 0.139-0.930, P=0.035) and high expression of PD-L1 (HR=0.155, 95%CI: 0.039-0.625, P=0.009) were protective factors for PFS. Most of the treatment-related adverse reactions in the whole group were grade 1-2, including bone marrow suppression (n=24), nausea (n=25), decreased appetite (n=24), fatigue (n=22), vomiting (n=18), abnormal liver function (n=17), blood creatinine increased (n=10), and so on. These were tolerated by the patients. Conclusions: NSCLC patients who failed EGFR-TKIs treatment can tolerate adverse reactions related to immunotherapy combined with chemotherapy and bevacizumab treatment. PFS is significantly prolonged in those aged≤65 years and those with moderate and high expression of PD-L1.

Targeted Next-Generation Sequencing Identifies Additional Mutations Other Than BCR-ABL in Chronic Myeloid Leukemia Patients: A Chinese Monocentric Retrospective Study

The discovery of tyrosine kinase inhibitors (TKIs) has drastically changed the treatment of chronic myeloid leukemia (CML) patients. However, a proportion of patients still show resistance or intolerance to TKI therapy. These patients often carry somatic variants related to myeloid malignancies, indicating that additional mutations other than BCR-ABL1 may lead to TKI treatment failure or disease progression. We retrospectively evaluated 151 CML patients receiving TKI therapy and performed next-generation sequencing (NGS) analysis of 22 CML patients to explore the mutation spectrum other than BCR-ABL1 affecting the achievement of molecular responses. We also performed a comprehensive evaluation of clinical characteristics, distinct TKI options and scoring systems in the correlation with the molecular response of TKI. In total, 161 variants were detected by NGS, most of which were nonsynonymous SNVs (Figure A). The coexistence pattern among high-frequency variants was quite intricate(Figure B). Among 25 genes with allele mutation frequency (VAF) ≥5%, the most frequently mutated gene was ASXL1 (40.9%), followed by KMT2C (27.3%), DIS3 (18.2%), ATM (13.6%), DNMT3A (13.2%) and NOTCH3 (13.6%) (Figure C). NOTCH3 and RELN mutations were only carried by subjects failing to achieve a major molecular response (MMR) at 12 months, suggesting CML patients with NOTCH3 and RELN mutations might have poor long-term treatment effects (Figure D, Figure E). The distribution frequency of ASXL1 mutations was higher in the group that did not achieve MR4.0 at 36 months (P=0.023), indicating that ASXL1 mutation was an adverse factor for the achievement of MR4.0. Furthermore, subjects in this group possessed more mutations related to activation signaling pathway genes (P=0.003) and chromatin modification genes (P=0.029) (Figure F). In the group that failed to achieve MR4.5 at 12 months, there were more subjects carrying >2 gene mutations (9.1% vs. 63.6%, P=0.024), implying that it is less likely to achieve MR4.5 with the increase in the number of mutated genes and that the existence of more than two mutations is a poor prognostic factor for achieving deep molecular response (DMR) (Figure G, Figure H). However, no statistical significance was found in the effects of mutations in ASXL1 (P=0.371), KMT2C (P=0.079), DIS3 (P=0.467), ATM (P=0.280), DNMT3A (P=0.479) and NOTCH3 (P=0.479) on PFS. In the analysis of clinical characteristics, multivariate analysis identified hemoglobin concentration (HB) (relative risk, [RR], 1.023; P=0.08) and BCR-ABLIS level at 12 months (RR, 2.485; P＜0.001) as independent predictive covariates for MR4.5(Figure I). Subjects who received initial second-generation TKI (2GTKI) were more likely to achieve early molecular response (EMR) at 3 months (P=0.012), BCR-ABLIS ≤1% at 6 months (P=0.012) and MMR at 12 months (P=0.018) than initial first-generation TKI (1GTKI). And there did not show statistical significance in achieving BCR-ABLIS≤1% at 6 months, MMR at 12 months and MR4.5 between original 1GTKI (Group A1, Glevic, Novartis, Switzerland) and generic 1GTKI (Group A2, Genike, Chiatai Tianqing Pharmaceutical Group Co., Ltd.). As for scoring system, ELTS score was sensitive in differentiating whether patients could achieve EMR (P=0.001) and MMR (P=0.004). EUTOS scoring system had a significant difference in achieving MMR at 12 months (P=0.049). However, there was no significant difference in subjects stratified by the Hasford and Sokal scoring systems. In conclusion, ASXL1 mutation, activation signal pathway genes, chromatin modification genes and the presence of >2 mutations were adverse factors in achieving molecular response with TKI treatment. HB and MMR were independent factors for DMR, and initial 2GTKI therapy was better than 1GTKI in the achievement of molecular response. For the scoring system, we found that the ELTS score was the best in predicting the efficacy of TKI therapy. Acknowledgement: The work was supported by the Key R&D Program of Zhejiang (No. 2022C03137.), Public Technology Application Research Program of Zhejiang, China (No. LGF21H080003.), the Key Project of Jinhua Science and Technology Plan, China (No. 2020-3-011.), the 2019-2022 Key Medical Discipline (Haematology) Fund of Jinhua, China. Correspondence to: Dr Jian Huang, The First Affiliated Hospital of Zhejiang University School of Medicine. No. 79 Qingchun Road. Hangzhou, Zhejiang, PR China. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

PB1895: ANALYSIS REVEALS CANCER-ASSOCIATED GENE MUTATIONS IN CHRONIC MYELOID LEUKEMIA PATIENTS ：A CHINESE MONOCENTRIC RETROSPECTIVE STUDY

Background: Recent genetic studies have shown that tyrosine kinase inhibitors (TKI) resistant or intolerant Chronic myeloid leukemia (CML) patients often carry one or more genetic changes related to myeloid malignancies, suggesting that additional mutations other than ABL1 mutation may lead to TKI treatment failure or disease progression. Aims: We performed sequencing of 22 CML patients using next-generation sequencing (NGS). 23 CML patients tested WT1 expression detection. Methods: The second cohort (No. 1) tested by second-generation sequencing included 22 CML patients. The second cohort (No. 2) tested by WT1 expression detection included 23 CML patients. The study has been approved by the institutional ethics committee and patients provided written informed consent in accordance with the Fourth Affiliated Hospital of Zhejiang University School of Medicine.Definitions Response was defined as BCR-ABL1 transcript levels according to the International Scale (IS) ≤10%, ≤1% and ≤0.1% at 3, 6 and 12 months after initiating TKI-therapy. Molecular response with 4.0 log reduction (MR4.0) was defined as 0.01%. Molecular response with 4.5 log reduction (MR4.5) was defined as 0.0032% or less BCR-ABLtranscript level. The comparison between groups of measurement data showed that the data with normal distribution were analyzed by one-way ANOVA or t-test. Results: Overall, a total of 51 mutations involving 25 genes were detected most frequently in the ASXL1 (40.9%), KMT2C (27.3%), DIS3 (18.2%), ATM (13.6%), DNMT3A (13.6%) and NOTCH3 (13.6%) genes. NOTCH3 and RELN mutations were only 3/20 (10%) in those who did not reach 12M-MMR group. Although the difference was not statistically significant (P = 0.214, P = 0.476). The number of median mutant genes in reaching MR4.5 group was 1 (0-3). The number of median mutant genes in not reaching MR4.5 group was 4 (0-6), and the difference was statistically significant (P = 0.033). ASXL1 mutation in reaching MR4.0 group and non reaching MR4.0 group was statistically significant (P = 0.023). With KMT2C mutation group had lower PFS (P = 0.079). There was a potential statistical difference between low WT1 expression group and high WT1 expression group in early EMR after a 3-month treatment (P = 0.089).The research was supported by the Key R&D Program of Zhejiang, No. 2022C03137; Public Technology Application Research Program of Zhejiang, China, No. LGF21H080003; the Key Project of Jinhua Science and Technology Plan, China, No. 2020-3-011; the 2019–2024 Academician Workstation of the Fourth Affiliated Hospi-tal of the Zhejiang University School of Medicine; the 2019–2022 Key Medical Dis-cipline (Hematology) Fund of Jinhua, China. *Correspondence to: Dr Jian Huang, Department of Hematology, The Fourth Affiliated Hospital of Zhejiang University School of Medicine. N1 Shangcheng Road. Yiwu, Zhejiang, Peoples R China.E-mail: [email protected] Image:Summary/Conclusion: Genomic risk assessment can improve the risk stratification of clinical diagnosis, and can more accurately identify patients with poor effect of TKI treatment in the era of TKI treatment. The treatment effect of CML patients with NOTCH3, RELN and ASXL1 mutations and more than two gene mutations is poor. CML patients with high expression of WT1 at diagnosis have poor therapeutic effect in the early stage.

Pembrolizumab for the better treatment of EGFR-mutant T790M-negative advanced lung adenocarcinoma patients than dual treatment of pemetrexed plus platinum after tyrosine kinase inhibitor treatment failure.

The treatment of lung cancer patients, especially those with epidermal growth factor receptor (EGFR)-mutant T790M-negative adenocarcinoma, after first- or second-line tyrosine kinase inhibitor (TKI) treatment failure is challenging due to the poor prognosis and limited effectiveness of platinum two-drug chemotherapy or chemotherapy plus anti-angiogenesis therapy. It is well-known that pembrolizumab monotherapy exhibits low toxicity and long-term survival, but it is unknown in these patients. From September 2018 to March 2021, 460 patients in Jiangmen Central Hospital were included and 82 patients with disease progression in lung adenocarcinoma who remained T790M-negative on the second biopsy were screened. Two groups were divided according to treatment status, and simple random sampling was performed to obtain 32 cases respectively. The safety of the patients was subsequently evaluated by telephone follow-up. The objective response rate (ORR) and disease control rate (DCR) in the pembrolizumab group were 15.63% and 53.13%. In the chemotherapy group, the ORR was 8.33% and the DCR was 25% (P<0.05). In the pembrolizumab group, the progression-free survival (PFS) [14.65 months, 95% confidence interval (CI): 13.03 to 16.28] was significantly higher than that of the control group (9.54 months, 95% CI: 8.43 to 10.65) (P<0.05). In the univariate analysis, programmed cell death protein 1 ligand (PD-L1) expression, smoking status, gender, and whether first-line chemotherapy was associated with survival. In the multivariate analysis, gender [P=0.001; hazard ratio (HR) 10.98, 95% CI: 2.49-46.67], first-line chemotherapy (P=0.037; HR 4.5, 95% CI: 1.1-4.81), and PD-L1 expression (P=0.039; HR 0.16, 95% CI: 0.04-0.68) were correlated with patient survival. Grade 3 or grade 4 treatment-related adverse events were not found in the pembrolizumab group, while 2 cases of grade 3 or 4 treatment-related adverse events occurred in the control group. In advanced lung adenocarcinoma patients with EGFR-mutant T790M-negative after TKI treatment, pembrolizumab had a higher ORR and PFS. Pembrolizumab in women with first-line chemotherapy and PD-L1 ≥25% of those patients may have a good response and a low rate of adverse reactions. A multicenter, prospective, evidence-based study of pembrolizumab salvage therapy in those patients is warranted for posterior line treatment.

A deep learning approach to discover cyclin-dependent kinases 12 (CDK12) inhibitors in breast cancer.

e15086 Background: Although anti-HER2 tyrosine kinase inhibitors (TKIs) have significantly prolonged survival and improved prognosis in HER2-positive breast cancer patients, resistance is a constant obstacle leading to TKIs treatment failure and tumor progression. Our previous research shows dual inhibition of HER2/CDK12 will prominently benefit the outcomes of patients with HER2-positive breast cancer by sensitizing the tumors to anti-HER2 TKIs treatment. Unfortunately, there is no high selective CDK12 inhibitor in clinical. This study sought to accelerate the CDK12 inhibitor discovery by a deep learning approach. Methods: We developed a large-scale self-supervised graph neural network to learn 11 million unlabelled compounds for molecular representation. And the model was trained with 233,823 kinase domain sequences by drug-target interaction datasets from the BindingDB database. Then, this transformer architecture was used to predict potential CDK12 inhibitors. The candidate hits compounds with diverse skeletons were divided into 50 clusters by clustering analysis. The representative hits with high transform scores and rational binding mode in each cluster were detected in the homogeneous time-resolved fluorescence CDK12 kinase assay. Candidate clusters with half-maximal inhibitory concentrations (IC50) lower than 200 nM were selected into expanded validation and kinase selective panel detection. xCELLigence RTCA system was adopted to monitor cell growth and sensitivity of breast cancer cells to CDK12 inhibitors in a real-time manner. Results: Our predictive transfer learning model yielded satisfactory predictions on various targets including CDK12. We screened a total of 4,527,236 compounds and recommended 50 clusters of potential CDK12 inhibitors for further enzyme activity detection. The IC50 of seven cluster representative compounds is lower than 10 uM. Expanded screening of 500 compounds in these clusters discovered a lot of hits with highly CDK12 selective inhibition. Half of the new compounds are showed promising proliferation inhibition in breast cancer cell lines. Conclusions: Our study provides a highly efficient and end-to-end deep learning approach to discover highly selective CDK12 inhibitors in breast cancer.

Burden of tyrosine kinase inhibitor failure in Chinese chronic myeloid leukemia patients: a systematic literature review.

Aim: To conduct a systematic literature review of real-world evidence on the burden of tyrosine kinase inhibitor (TKI) failure in Chinese patients with chronic myeloid leukemia (CML). Methods: We identified 155 references in Chinese- and English-language journals from 2001 to 2021. Results: The age-adjusted mortality rate in Chinese CML patients was decreasing. Imatinib treatment had a higher annual treatment failure risk than nilotinib (0.199 vs 0.041). Patients with TKI treatment failure tended to be young (median: 38.6years), have progressive disease (44.3%) and harbor BCR-ABL1 mutations (51.6%). The disease burden of TKI treatment failure included reduced health outcomes and increased health resource utilization and costs. Conclusion: CML relapse cases could continuously rise in China due to increasing TKI treatment failure over extended survival.

Analysis of the Sub-Clonal Origins of Compound Mutations in Patients with Refractory Ph+ Malignancies Treated with Ponatinib

Background: Ponatinib is a tyrosine kinase inhibitor (TKI) with potent activity against BCR-ABL1 and all single resistance mutants. In the Phase 2 PACE study (NCT01207440), ponatinib induced high rates of major hematological response (MaHR) and major cytogenetic response (MCyR) in BP-CML (31% and 23%) and Ph+ ALL (41% and 47%) patients (pts), even though 91% of pts had received at least 2 prior TKIs. However, the median progression free survival was only 4 months for BP-CML and 3 months for Ph+ ALL (Cortes JE, et al. N Engl J Med 2013). Previous studies have shown that certain compound mutations can cause resistance to ponatinib (Zabriskie et al. Cancer Cell 2014), but such mutations were found to be extremely rare in CP-CML pts from the PACE trial despite being heavily pre-treated (Deininger MW, et al. Blood 2016). Here we utilize a multi-level sequencing strategy that combines Sanger Sequencing (SS), Next Generation Sequencing (NGS), and single molecule Duplex Sequencing (DS), which is orders of magnitude more sensitive than NGS, to profile the mutational mechanisms that may account for the survival outcomes observed in BP-CML/Ph+ ALL pts treated with ponatinib in PACE.Methods: The BCR-ABL1 mutation status of the 267 CP-CML, 62 BP-CML, and 32 Ph+ ALL pts treated in the PACE trial was examined using SS and/or NGS as described previously (Cortes JE, et al. N Engl J Med 2013; Deininger MW, et al. Blood 2016). Mutations detected by SS at the end of treatment (EOT) but not at baseline (BL) were considered to have been gained during ponatinib treatment. DS was performed to determine whether mutations gained on treatment could be found in small sub-populations of cells at BL and to provide a direct measurement of the mutation burden in the ABL1 gene. Two distinct mathematical models were used to predict the size of pre-existing resistant sub-populations in BP-CML/Ph+ ALL pts based on the overall ABL1 mutation burden and the doubling time of clonal outgrowth.Results: The number of BCR-ABL1 mutations in BP-CML/Ph+ ALL patients at BL, as detected by SS, was not a significant predictor of intrinsic resistance to ponatinib in refractory patients with Odds Ratios (OR) of 1.0 (0.2-4.9 95% CI) for MaHR and 2.4 (0.5-12.5 95% CI) for MCyR. However, a larger proportion of BP-CML/Ph+ ALL pts, 30/61, acquired mutations in BCR-ABL1 at EOT vs. 19/130 CP-CML pts (p<0.005, OR 7.7 [1.5-53.3 95% CI]). Nearly all BP-CML/Ph+ ALL, but not CP-CML, pts who acquired mutations had BCR-ABL1 mutations at BL detected by SS: 27/30 vs. 10/19 CP-CML pts (p=8.5E-7, OR 5.6 [2.7-12.1 95% CI]). At least 19/27 BP-CML/Ph+ ALL pts had compound mutations at EOT, including 2 pts with 3 mutations in the same allele. Since the presence of pre-existing mutations from earlier TKI therapy is associated with further gain of mutations on ponatinib treatment, we used ultra-sensitive DS to determine if these resistant clones could be detected at BL. In 3/18 pts tested, the resistant clone was directly detected at BL in a small minority of cells. Mathematical modeling of leukemia growth rates in pts with mutation gains predicts that pre-treatment resistant population sizes range from ~1 cell to 2.7 x 108 cells in the peripheral blood, out of a total population of leukemic cells that ranged from ~1010 to 5 x 1011. Furthermore, DS and mathematical modeling allowed us to predict that many, if not all, documented resistance mutations were pre-existing in most refractory BP-CML/Ph+ ALL pts treated with ponatinib in the PACE trial.Conclusions: Clonal BL BCR-ABL1 mutation status determined by SS does not predict intrinsic resistance to ponatinib. However, in contrast to CP-CML, BP-CML/Ph+ ALL pts are more likely to relapse on ponatinib with acquired mutations in BCR-ABL1. Two distinct mathematical modeling strategies suggest that most, if not all, BCR-ABL1 resistance mutations in BP-CML/Ph+ ALL pts that emerge during ponatinib treatment were pre-existing. The majority of these pre-existing mutations were compound mutations whose evolutionary path was likely determined by prior TKI therapy. These findings emphasize the risk of the large population expansions that occur following prior TKI treatment failure and may explain the relatively high 2-yr event free survival rate of 81% observed in previously untreated Ph+ ALL pts treated with ponatinib (+hyper-CVAD) (Jabbour E, et al. Lancet Oncol 2015, Sasaki, et al. Cancer 2016).Study sponsor: ARIAD Pharmaceuticals, Inc. DisclosuresPritchard:ARIAD: Employment, Equity Ownership. Schmitt:TwinStrand: Equity Ownership. Hodgson:ARIAD: Employment, Equity Ownership. Lustgarten:ARIAD: Employment, Equity Ownership. Tang:ARIAD: Employment, Equity Ownership. Loeb:TwinStrand: Equity Ownership. Radich:Novartis: Consultancy, Other: laboratory contract; Bristol-MyersSquibb: Consultancy; Pfizer: Consultancy; ARIAD: Consultancy; TwinStrand: Consultancy. Rivera:ARIAD: Employment, Equity Ownership.

Abstract 4663: Combination of c-Met inhibitor tepotinib (MSC2156119J) and a third-generation EGFR inhibitor can overcome double resistance mediated by EGFR T790M mutation and c-Met amplification in non-small cell lung cancer models

Abstract In clinical practice, acquired resistance to EGFR tyrosine kinase inhibitors (TKIs) in EGFR-mutant non-small cell lung cancer (NSCLC) causes EGFR TKI treatment failure. Several resistance mechanisms have been identified; EGFR T790M mutation and aberrant activation of c-Met through c-Met gene amplification and/or HGF upregulation are the two most common mechanisms. We have previously shown that combining the highly selective c-Met inhibitor tepotinib with 1st or 2nd-generation EGFR TKIs overcomes gefitinib resistance caused only by c-MET amplification. Third-generation EGFR TKIs such as rociletinib have been developed specifically to overcome T790M-mediated resistance to EGFR TKIs. We hypothesized that combining a 3rd-generation EGFR TKI with tepotinib could overcome resistance to a 1st or 2nd-generation EGFR TKI (e.g. gefitinib) regardless of the mechanism of resistance. We assessed the efficacy of tepotinib and a Merck-synthesized version of rociletinib (MSR) as monotherapy or in combination in a NSCLC cell line xenograft model (HCC827-GR-T790M) with an EGFR del 19 mutation, c-Met amplification, and exogenous T790M expression, and in a patient-derived xenograft model (DFCI081) with an EGFR del 19 mutation and c-Met amplification, but not a T790M mutation. Single-agent tepotinib, erlotinib, and afatinib did not show antitumor activity in HCC827-GR-T790M xenografts and tumors progressed on treatment (median tumor volume [TV] changes &amp;gt;73%). The combinations of tepotinib + erlotinib and tepotinib + afatinib delayed tumor regrowth more than single-agent therapy, but were no more effective than single-agent MSR. In contrast, the combination of tepotinib + MSR had strong antitumor activity in this model with double resistance due to c-Met amplification and exogenous T790M expression, resulting in complete tumor regression (median TV change -97%). In the patient-derived xenograft model DFCI081, tumors progressed under single-agent therapy with the EGFR TKIs erlotinib, afatinib, and MSR, as expected given that this cell line expresses c-Met. In contrast, tepotinib induced complete tumor regression as monotherapy and when combined with each of the EGFR TKIs (median TV change -100%). Complete regression was maintained until the end of the 60-day observation period after treatment termination. These findings are compatible with the hypothesis that combining tepotinib with a 3rd-generation EGFR TKI to overcome double resistance to 1st and 2nd-generation TKIs mediated by T790M and c-Met in NSCLC is more effective than using either agent alone. Furthermore, whereas EGFR TKIs were not effective in tumor models showing c-Met amplification and EGFR del 19 mutation but not T790M mutation, ie resistance is c-Met mediated, tepotinib monotherapy induced complete tumor regression. Citation Format: Manja Friese-Hamim, Giuseppe Locatelli, Friedhelm Bladt. Combination of c-Met inhibitor tepotinib (MSC2156119J) and a third-generation EGFR inhibitor can overcome double resistance mediated by EGFR T790M mutation and c-Met amplification in non-small cell lung cancer models. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4663.

Increasing economic burden of tyrosine kinase inhibitor treatment failure by line of therapy in chronic myeloid leukemia

Objective:To assess the economic burden of tyrosine kinase inhibitor (TKI) treatment failure in chronic myeloid leukemia (CML), by assessing all-cause health care resource use (HCRU) and costs in the year after treatment failure by line of therapy (LOT; 1L/2L/3L) using real-world data.Methods:Treatment episodes initiating a TKI of interest (index TKI) during June 2008–December 2011 were identified from the IMS PharMetrics Plus Health Plan Claims Database for adult patients with CML diagnosis (ICD-9-CM 205.1x), 120 days pre-index continuous enrollment (CE) and no clinical trial participation. Episodes experiencing treatment failure, defined as switch to a non-index TKI or discontinuation of index TKI (gap of ≥ 60 days), and with 1 year CE post-failure, were analyzed. LOT was determined by number of unique TKIs used in the pre-index. All-cause HCRU and costs (2012 USD) in the 1 year post-failure were assessed by LOT, and the comparisons between 1L and 2L failures were also adjusted using multivariate generalized linear models (GLMs) to control for underlying differences.Results:A total of 706 episodes were identified (518 1L; 180 2L; 8 3L). Unadjusted HCRU over 1 year post-failure increased significantly. This was accompanied by a significant increase in unadjusted mean costs for 2L failures vs. 1L failures ($99,624 vs. $78,667, p = 0.021, Δ$20,957). Following the adjustment using GLMs, adjusted mean costs were 38% higher (95% CI 1.14–1.68), driven primarily by use of medical services. In adjusted analyses, compared to 1L, 2L failures had: 45% more ambulatory visits (mean 31 vs. 21, 95% CI 1.26–1.66), 75% higher risk of hospitalization (33% vs. 23% hospitalized, 95% CI 1.16–2.64), and 73% higher medical costs (95% CI 1.31–2.29). Medical costs comprised a greater proportion of total costs in 2L vs. 1L (55% vs. 44%); pharmacy costs did not increase significantly.Conclusions:The economic burden over 1 year post TKI failure increased with each sequential line of TKI treatment failure.

Economic Burden of Tyrosine Kinase Inhibitor Treatment Failure in Chronic Myeloid Leukemia.

The economic burden of tyrosine kinase inhibitor (TKI) treatment failure in chronic myeloid leukemia (CML) is not well understood. The objective of this study was to quantify the economic burden associated with treatment failure versus successfully remaining on TKI therapy. Treatment episodes for adult CML patients initiating a TKI of interest (imatinib, dasatinib, or nilotinib; index TKI) during July 1, 2008, to December 31, 2011, with continuous enrollment for≥ 120 days before and 1 year after the initiation were identified from the IMS PharMetrics Plus Health Plan Claims Database. Eligible episodes of TKI treatment failure were matched to those without failure using propensity scores based on patients' baseline demographic and clinical characteristics. Treatment failure was defined as a switch to a nonindex TKI or discontinuation (gap in pharmacy claims≥ 60 days) of index TKI over the 1-year follow-up. Mean all-cause health care resource utilization and costs per episode (in 2012 US dollars) over follow-up were compared between failures and nonfailures. Among 1774 eligible episodes, 547 failures were matched to 547 nonfailures. Failures had fewer TKI prescription fills but higher utilization of all other services versus nonfailures. Consequently, failures incurred lower pharmacy costs ($51,238 vs. $72,450; Δ-$21,212) but higher medical costs ($52,619 vs. $18,180; Δ$34,439) than nonfailures, resulting in higher total costs ($103,857 vs. $90,630; Δ$13,227) (all P< .05). Total health care costs are higher for episodes of TKI treatment failure than those of ongoing treatment, largely as a result of costly medical (nonpharmacologic) services. Avoiding treatment failure by optimal CML management may reduce health care costs.

Emergence of BCR-ABL kinase domain mutations associated with newly diagnosed chronic myeloid leukemia: a meta-analysis of clinical trials of tyrosine kinase inhibitors.

Tyrosine kinase inhibitors (TKIs) are a mainstay of treatment for patients suffering from chronic myeloid leukemia (CML). Testing for various mutations in the BCR-ABL gene may help predict lack of response to specific TKIs where resistance has developed. To estimate the emergence of BCR-ABL kinase domain mutations associated with newly diagnosed CML patients exposed to first-line TKI treatment. Published studies were identified using a structured search of online databases. Original research studies were included if they reported the incidence of specific BCR-ABL kinase domain point mutations after first-line TKI treatment failure or baseline mutations for second-line TKI treatment following first-line treatment failure. Meta-analysis of mutation rates across studies was based on DerSimonian and Laird's random-effects model. Of 1,323 citations, 12 studies met the inclusion criteria, involving a total of 1,698 patients. Overall mutation rates (95% CI) were imatinib 9.7% (6.2%-13.3%); dasatanib 1.7% (0.0%-4.3%); and nilotinib 3.3% (0.0%-7.7%). The most common specific mutations were T315I, E255K, and M351T. T315I mutations constituted 58% (7 of 12) of dasatinib-related mutations and 13% (15 of 117) of imatinib-related mutations. Lack of response to TKIs associated with mutation in the BCR-ABL gene was significantly higher in imatinib-treated patients, and all mutations arose after treatment. T315I was a common treatment-emergent mutation. Further research is needed to assess the prognostic value of testing for mutations and the economic implications of treatment-emergent mutations.