Tyrosine Kinase Inhibitors Treatment Research Articles

Brain metastasis, EGFR mutation subtype and generation of EGFR-TKI jointly influence the treatment outcome of patient with EGFR-mutant NSCLC

Non-small cell lung cancer (NSCLC) with epidermal growth factor receptor (EGFR) mutation is brain metastasis (BM)-prone. We determined the impact of this hallmark, along with EGFR subtype and generation of tyrosine kinase inhibitor (TKI) treatment, on patients’ outcome. 553 metastatic EGFR-mutant NSCLC patients received front-line EGFR-TKI treatment. Progression-free survival (PFS), overall survival (OS) and secondary T790M rate were analysed. BM was observed in 211 (38.2%) patients. BM (HR 1.20 [95% CI 0.99–1.48]; p = 0.053), ECOG PS 0–1 (HR 0.71 [95% CI 0.54–0.93]; p = 0.014) and afatinib treatment (HR 0.81 [95% CI 0.66–0.99]; p = 0.045) were associated with PFS. Afatinib-treated patients without BM demonstrated a significantly longer PFS (16.3 months) compared to afatinib-treated patients with BM (13.7 months) and to gefitinib/erlotinib-treated patients with (11.1 months) or without BM (14.2 months; p < 0.001). CNS-only progression trended higher in afatinib-treated patients. ECOG PS 0–1 (HR 0.41 [95% CI 0.31–0.56]; p < 0.001) and EGFR L858R mutation (HR 1.46 [95% CI 1.13–1.88]; p = 0.003), but not BM, were the predictors for OS. BM (OR 2.02 [95% CI 1.02–4.08]; p = 0.040), afatinib treatment (OR 0.26 [95% CI 0.12–0.50]; p < 0.001) and EGFR L858R mutation (OR 0.55 [95% CI 0.28–1.05]; p = 0.070) were associated with secondary T790M rate. In BM patients, gefitinib/erlotinib-treated ones with 19 deletion mutation and afatinib-treated ones with L858R mutation had the highest and the lowest T790M rate (94.4% vs. 27.3%, p < 0.001), respectively. BM and generation of EGFR-TKI jointly impact PFS and secondary T790M rate in patients with EGFR-mutant NSCLC, whereas OS was mainly associated with EGFR subtype.

MODL-05. ELUCIDATING THE EPIGENETIC DRIVEN TRANSCRIPTIONAL PROGRAM UNDERLYING THE INVASIVE PHENOTYPE OF THE NOVEL EGFRA289VARIANT OF GBM

Abstract Glioblastoma (GBM) is a devastating malignant brain tumor with a dismal 15-month median survival. Current treatments have not been successful at preventing recurrence, thus novel therapeutic modalities are urgently needed. EGFR is a tyrosine kinase receptor that is commonly mutated in ~60% of GBM and is an attractive target for GBM precision oncology. Aberrant EGFR signaling drives tumor proliferation and maintenance, thus further study of common EGFR alterations in GBM, including interstitial deletions (EGFRvIII) and missense mutations (EGFRA289), is needed to develop effective EGFR-targeted therapies. This is especially important in the context of a retrospective study that showed EGFRA289 driven GBM was more clinically aggressive. EGFRA289 positive GBM had a unique, invasive radiographic signature on magnetic resonance imaging (MRI) and patients had a significantly shorter median overall survival of 6-months. We hypothesize that the malignant clinical phenotypes of EGFRA289 GBM are driven by unique epigenomic and transcriptomic, collectively referred to as (epi)genomic, programs linked to EGFRA289 signaling. Prior studies using established cell lines show that EGFRvIII overexpression drives malignant (epi)genomic programs mediated by transcription factor (TF) activation. To better understand variant specific EGFR biology in GBM, we have developed novel genetically engineered mouse astrocyte (mAc) models that overexpress wtEGFR (MA-WT), vIII (MA-v3), and EGFRA289V (MA-A289V). Phospho-immunoblots show that these mAc models have differential EGFR ligand dependance, which match published studies. Despite this difference in EGFR signaling, all mAc models have similar sensitivities to EGFR tyrosine kinase inhibitor (TKI) treatment. However, MA-A289V is more proliferative and migratory compared to the other mAc variants. Preliminary RNA-Seq on our mAc models has identified similar and unique TFs differentially expressed (DE) between MA-v3 and MA-A289V when compared to parental MA-C. Integration of RNA-Seq and CUT&amp;RUN followed by mechanistic genetic interrogation will elucidate the relationship between EGFR variant-specific (epi)genomic mechanisms and their unique malignant phenotype.

MODL-31. USING GENETICALLY-ENGINEERED MOUSE MODELS OF TREATMENT-NAÏVE AND -EXPERIENCED NTRK FUSION-DRIVEN GLIOMAS FOR PRE-CLINICAL TRIALS OF TYROSINE KINASE INHIBITORS

Abstract Gene fusions that contain the kinase domain of the three neurotrophic tyrosine kinase receptor (NTRK) genes are frequently observed in infantile high and low-grade gliomas. While these tumors initially respond to treatment with NTRK tyrosine kinase inhibitors (TKIs), treatment resistance and recurrence eventually occur. Treatment-resistant tumors frequently harbor either on-target (in the fusion's kinase domain) or off-target mutations (in downstream effectors of mitogen-activated signaling pathways). Therefore, overcoming this resistance requires a three-pronged strategy: 1) identifying new kinase inhibitors capable of targeting NTRK fusions with on-target kinase domain mutations, 2) identifying druggable targets in the downstream pathways activated by these fusions, and 3) targeting persister cells that contribute to disease recurrence. We utilized the RCAS/tv-a system to establish several mouse models of NTRK1/2/3 fusion-driven gliomas. We found that all analyzed fusions were strong oncogenic drivers on their own, while the loss of additional tumor suppressors (Cdkn2a or Pten) further increased the aggressiveness of these tumors. These mouse models exhibit similar treatment responses as human tumors, initially responding to various TKIs but eventually recurring due to the presence of treatment-resistant persister cells. To mimic the emergence of resistance-associated mutations, we directly introduced different kinase domain mutations into the NTRK fusion sequence or co-expressed secondary mutations (BRAF-V600E or KRAS-G12D). These modifications altered the response to TKI treatment both in vitro and in vivo. Overall, these models serve as valuable tools for studying the biology of NTRK fusion tumors, their treatment response in a treatment-naïve or -experienced state, and the mechanisms underlying treatment resistance.

CTIM-23. SAFETY RUN-IN RESULTS OF A PHASE I/II STUDY TO EVALUATE ATEZOLIZUMAB IN COMBINATION WITH CABOZANTINIB IN PATIENTS WITH RECURRENT GLIOBLASTOMA

Abstract BACKGROUND Immune checkpoint inhibitor therapy has shown limited efficacy in the treatment of glioblastoma (GBM) in both newly diagnosed and recurrent disease which underscores the need for rational combinatorial strategies. Atezolizumab is a humanized monoclonal antibody that targets PD-L1 and has been shown to enhance the magnitude of tumor-specific T cell responses. Combining PD-L1/PD-1 axis inhibition with targeted therapy like cabozantinib which has both anti-tumoral and immunomodulatory properties may result in synergistic effects. Cabozantinib may modulate the myeloid cell population to overcome resistance to anti-PD-1/PD-L1 therapy. METHODS Eligibility criteria were those with recurrent GBM, age &amp;gt;18 yrs with 1st or 2nd recurrence who were naïve to antiangiogenic and receptor tyrosine kinase inhibitor treatment. The primary objective of the phase I run-in was to evaluate the safety of atezolizumab in combination with cabozantinib. RESULTS 1 dose-limiting toxicity (DLT) of a grade 3 ALT elevation was observed in the first 6 DLT evaluable patients treated during the safety run-in phase. Regarding immune related adverse events, 1 of 6 patients developed atezolizumab-related grade 1 hypothyroidism. 1 of 6 patients developed study treatment-related grade 2 diarrhea. 2 patients after completion of 2 cycles of treatment had evidence of a partial radiographic response ( &amp;gt; 50% decrease). 1 patient had a near partial radiographic response (45% decrease). 2 patients had stable disease after completion of 2 cycles of the combinatorial treatment. 1 patient had only completed the first 28-day cycle, DLT period, and had not yet had imaging to reassess disease status. CONCLUSIONS The combination of atezolizumab with cabozantinib was tolerable, and no new safety signals were noted. In this small cohort of recurrent GBM patients, radiographic responses were observed. The phase II component of the trial is recruiting patients (n = 45) to evaluate clinical efficacy. Multi-omic correlative studies and fecal microbiome analyses are planned.

Moyamoya disease presenting with symptomatic ischemic stroke during new-generation tyrosine kinase inhibitor treatment: two illustrative cases.

Tyrosine kinase inhibitors (TKIs) have been widely used to treat chronic myeloid leukemia. Nilotinib and ponatinib, which are second- and third-generation TKIs, have been reported to cause cerebrovascular arterial complications. Here, we present two cases of moyamoya disease presenting with symptomatic ischemic stroke during new-generation TKI treatment. We judged that new-generation TKI treatment was a factor in symptomatic ischemic stroke of unknown moyamoya disease in both cases. Noninvasive examinations using magnetic resonance imaging or carotid ultrasonography should be performed before and during new-generation TKI treatment in order to prevent symptomatic ischemic stroke.

Somatic Mutations in Cancer-Related Genes Were Observed More Frequently in AYA CML Patients Compared to Elderly at Diagnosis, Whereas the Frequency Was Markedly Higher in Elderly Patients during TKI Treatment

CONCLUSIONS Several works showed that patients with chronic myeloid leukemia (CML) diagnosed at the age of 18 to 39 years (adolescent and young adults, AYAs), had a worse outcome of tyrosine kinase inhibitors (TKI) therapy compared to elderly patients. Only limited data on the molecular background differing AYA from the other age groups are available. To characterize the somatic mutation landscape of AYA CML patients in chronic phase at the time of diagnosis and follow-up in comparison to pediatric and elderly patients, and to follow their clonal evolution. Ph-positive acute lymphoblastic leukemia (Ph+ ALL) patients were included for comparison. Samples from the time of diagnosis (CML children N=18; AYA N=69; elderly N=70; Ph+ ALL children N=32, AYA N=13, elderly N=39) and TKI treatment follow-up (CML AYA N=69, elderly N=70) were analyzed with custom panel (22 whole genes, 40 selected exons). To follow the clonal evolution vital cells available from 10 CML patients were FACS sorted into populations of neutrophils, monocytes, T- and B-lymphocytes. Genomic BCR::ABL1 (g BCR::ABL1) and mutations were quantified using ddPCR in the sorted cells. At the time of diagnosis, a slightly higher frequency of somatic mutations was detected in AYA CML patients (28%, 19/69) compared to elderly (20%, 14/70) and children (11%, 2/18). The similar trend with markedly higher differences was observed in Ph+ ALL patients (62% 8/13 of AYA, 31% 12/39 of elderly, 22% 7/32 of pediatric). The most frequently mutated genes in AYA and elderly CML patients were identically ASXL1, DNMT3A, and TET2. Other mutated genes in AYA were FBXW7, IDH2, IKZF1 PHF6 and KDM6A, SETD2, SF3B1 in elderly. Considering the treatment response, a higher number of mutations at diagnosis were found in the AYA patients that responded optimally in follow-up (21%, 7/33) than in elderly 4% (1/26) and a slightly higher frequency of mutations in AYA patients who later failed on TKI (35%, 8/23) than in elderly 30% (13/43). In contrary, the frequency of acquired mutations or persisted mutations from diagnosis detected during TKI therapy was higher in elderly that failed on TKI therapy (67%, 29/43) than in AYA (52%, 12/23). The treatment failure was associated with de novo mutations in ABL1, RUNX1, TET2, ASXL1 in AYA and elderly CML patients, and in addition in IKZF1, and in DNMT3A, WT1, JAK2, respectively. Sorted cells were analyzed from the time of optimal response in 7 CML patients (+ diagnostic sample of 2 patients) with mutated ASXL1 in 6 patients, FBXW7 in 1 patient, DNMT3A in 1 patient and IDH2 in 1 patient. Samples from the time of therapy failure were available for sorting of 3 patients (+ diagnostic sample of 1 patient). All mutations were detected in g BCR::ABL1 positive neutrophiles and monocytes. The g BCR::ABL1 levels matched exactly with the levels of mutations except for 1 patient with ASXL1-mutated subclone (VAF 5%), which was found in g BCR::ABL1 positive cells. Similar data were found in patients with therapy failure. However, in one patient with secondary failure, the ASXL1 mutation detected at diagnosis disappeared. In all 10 CML patients, g BCR::ABL1 and mutations were not detected in T- and B- lymphocytes under the therapy. In diagnostic samples some g BCR::ABL1 and mutations positive T- and B- lymphocytes were found, but the levels were 3-4 log lower compared to levels in neutrophils and monocytes, which likely correspond to the contamination (purity check was done after sorting) than to the presence of a very small subset of physiological lymphoid cells. Of all age groups, AYAs had the highest number of mutations detected at the time of diagnosis with differences in the spectrum between CML and Ph+ ALL. A markedly higher de novo acquisition of mutations was observed in elderly CML patients during TKI treatment. Moreover, therapy failures were observed more frequently in elderly than in AYA CML patients. Clonal analysis of 10 CML patients outlined that myeloid, but not lymphoid stem/progenitor cells were mutated. Supported by MZCR NU21-07-00225

Benefit of TKI-Treatment in CML after Failing Molecular or Cytogenetic Milestones

No information is available on long-term survival of tyrosine kinase inhibitor (TKI)- treated patients failing the response milestones as defined by the European LeukemiaNet (ELN). Such information would help with decision making in failing patients when risks associated with comorbidities, drug toxicities or transplantation preclude switching to another TKI or other treatments. To systematically analyze survival after reaching, or not reaching, ELN-response milestones, 1342 imatinib-treated patients from CML-study IV with newly diagnosed CML in chronic phase and regular molecular and cytogenetic tests were studied. Survival was analyzed by landmark analyses at 3, 6, 12 and 24 months and followed up to 14 years. In patients who failed the failure milestones (&amp;gt;10% BCR::ABL1 IS at 6 months, &amp;gt;1% BCR::ABL1 IS at 12 months), 10- to 12-year survival ranged around 70%, 10-20% less than in responding patients. An example is shown in Fig. 1a which depicts survival of patients reaching, or not reaching 1% BCR::ABL1 IS at 12 months. Similar survival curves were observed for the other landmarks that indicate failure. Inbetween response rates of &amp;gt;1%-10% BCR::ABL1 IS at 12 months indicated survival probabilities similar to those of patients failing 1% BCR::ABL1 IS, whereas patients reaching &amp;gt;1%-10% BCR::ABL1 IS at 24 months showed a survival similar to that of patients failing 10% BCR::ABL1 IS. Switching to alternative therapies, mostly to dasatinib and nilotinib, was observed in 372 patients (26.9%) and did not change the main results. Although cytogenetic tests were increasingly replaced by molecular tests in the course of the study, sufficient cytogenetic analyses were available for comparison with molecular analyses. Complete and partial cytogenetic remissions have been shown to be roughly equivalent to 1% and 10% BCR::ABL1 IS, respectively. Wheras reaching, or not reaching cytogenetic and molecular reponse markers at 6 and 12 months was equivalent, not reaching a complete or partial cytogenetic remission at 24 months identified a small subgroup of patients (5.9%) with inferior survival (Fig. 1b). Age (more or less than 60 years) had no major impact on survival differences, but on hazard ratios for early death. Hazard ratios at 10% BCR::ABL1 IS by 6 months were 4.0 (95%CI:2.0-8.1) for the younger patients, but only 0.9 (95% CI: 0.4-1.8) for the elderly. The reasons for early death in the younger patients were mostly CML-related, whereas the elderly primarily died of other reasons. The data show that TKI-treated patients not reaching failure milestones still may derive benefit from continuing TKI-treatment and provide a basis for individualized decisions, if failing patients are confronted with risks of alternative treatments.

Full Treatment-Free Remission Outcome in Patients with Chronic Myeloid Leukemia in Chronic Phase Following One Year of Nilotinib De-Escalation: 96-Week Update of Dante Study

Introduction: Treatment-free remission (TFR) in chronic myeloid leukemia (CML) is achievable and recommended for patients (pts) in sustained deep molecular response (DMR) who can discontinue tyrosine kinase inhibitor (TKI) treatment while maintaining responses. Studies investigating TKI de-escalation before TFR are mainly in imatinib-treated CML chronic phase (CP) pts and suggest that this stepwise approach is feasible. Although nilotinib (NIL) is the first and only TKI approved for stopping therapy, nilotinib-based TFR optimization strategy has not been formally studied. Our phase II, single-arm, prospective, multicenter DANTE study (NCT03874858) aims to shed light on frontline NIL de-escalation in terms of full treatment-free remission (FTFR) (primary endpoint) and TFR rate (secondary endpoint) 96 weeks after the start of consolidation in CML-CP Italian patients. Our previously analysis on 40 pts with sustained DMR who entered in the TFR phase showed that approximately 68% of pts remained disease-free 1 year after stopping NIL 1. Here we present the data of 89 patients who successfully completed NIL de-escalation and at least 1 year of TFR. Methods: Adults patients with CML-CP treated with NIL 300 mg twice daily (bid) in first line for ≥3 years who achieved sustained DMR for ≥1 year (≥MR 4.0; BCR-ABL level ≤0.01% IS) were enrolled in 27 centers. The study consisted of 4 phases: screening (week [wk] −4-0), consolidation (wk 0-48), TFR (wk 48-144), and follow-up (until wk 144). Ongoing treatment with ≥400 mg/day dose was allowed at study entry. During consolidation pts were treated with NIL 300 mg once daily (qd). At the end of consolidation phase, pts with sustained DMR entered TFR phase and NIL was discontinued. Patients with major molecular response (MMR [BCR-ABL ≤0.1% IS]) but without sustained DMR continued NIL 300 mg qd. At any time, pts with loss of MMR returned to NIL 300 mg bid. During TFR phase, BCR-ABL levels were monitored monthly from wk 52-96 and then every 3 months. Here we report updated results collected until the cut-off date of 8th February 2023, when all patients who entered the TFR phase had completed at least one year of TFR or switched to full-dose phase or entered in survival follow-up or discontinued the study. The primary endpoint of the study is FTFR, defined as the percentage of patients (in both TFR phase and NIL half-dose) in MMR or better at week 96. Results: Overall, 113 pts were screened and 107 entered in consolidation phase. The baseline characteristics and the clinical outcome of the 107 patients are shown in Tab 1. Specifically, 98/107 (92%) completed the consolidation phase, while 9 pts discontinued permanently (2 due to loss of MMR, 3 due to adverse events (AEs) and 4 due to patient decision). Noteworthy, 89/107 (83%) with sustained DMR entered in the TFR phase, while 9 pts maintained MMR but not sustained DMR (7 entered in NIL half dose follow-up period and were in FTFR at 96 weeks, while 2 discontinued the study before week 96). At 1 year cut-off data, 71/107 pts (66.4%) were in FTFR, meeting the primary endpoint of study. We observed 64/89 pts (72%) who entered in the TFR phase remained in MMR or better after a median duration of 20.4 months (interquartile range [IQR] 1.9-32.4), meeting the main secondary endpoint of the study. The median time to loss of MMR in TFR phase was not reached (Fig 1). The remaining 25/89 pts (28%) had discontinued the TFR phase (21 pts due to loss of MMR, 1 due to AEs and 3 pts for subject/guardian decision). The median time spent in TFR phase until loss of MMR was 3.6 months (IQR 2-19 months). At the cut-off data, 23/24 pts (96%) who restarted NIL after MMR loss regained MMR and MR4 after a median time of 2.7 months (IQR 1.0-5.6) and 4.1 months (IQR 1.2-10), respectively. Notably, 20/24 (83%) who restarted NIL after MMR loss achieved MR4.5 after a median time of 6.4 months (IQR 2.0-16.9). During the TFR phase, all-grade AEs were reported in 64/89 pts (72%). One pt (1.1%) had serious AE (pneumonia). No deaths and disease progressions were observed. Conclusions: Our findings show that 66.4% pts who entered in consolidation phase achieved FTFR 96 weeks after the start of the consolidation period, and approximately 72% of pts remained disease-free at 1 year after stopping NIL. Overall, our data suggest that de-escalation of NIL before TFR attempt in CML-CP pts with sustained DMR can be a successful dose-optimization strategy. 1Stagno F. et al., Blood (2022) 140 (Supplement 1): 9614-9616.

Lisocabtagene Maraleucel (liso-cel) in R/R CLL/SLL: 24-Month Median Follow-up of TRANSCEND CLL 004

Background:Patients with R/R CLL/SLL who experience intolerance to or disease progression after Bruton tyrosine kinase inhibitor (BTKi) and venetoclax treatment have no established standard of care and poor outcomes, indicating a critical unmet need. Liso-cel, an autologous, CD19-directed, 4-1BB CAR T cell product, has demonstrated efficacy in large B-cell lymphoma and CLL/SLL. In the primary analysis of the phase 1/2, single-arm, multicenter TRANSCEND CLL 004 (NCT03331198) study, a single administration of liso-cel demonstrated rapid, deep, and durable responses and a manageable safety profile in patients with R/R CLL/SLL, including those with progression on previous BTKi and venetoclax failure (Siddiqi T, et al. Lancet 2023). The primary endpoint was met in the prespecified subset of efficacy-evaluable patients with disease progression on BTKi and venetoclax failure (primary efficacy analysis set [PEAS]) at a target dose of 100 × 10 6 CAR + T cells (null hypothesis: ≤ 5%) with the rate of complete response/remission (CR) and CR with incomplete marrow recovery (CRi) by an independent review committee (IRC) per 2018 International Workshop on CLL (iwCLL) criteria at 18.4% ( P = 0.0006). Here we report results from TRANSCEND CLL 004 with a median follow-up of 23.5 months. Methods: Patients must have received ≥ 2 prior lines of therapy, including a BTKi. Eligible patients received liso-cel at a target dose of either 50 × 10 6 (dose level [DL] 1) or 100 × 10 6 (DL2) CAR + T cells after lymphodepleting chemotherapy. The primary endpoint was CR/CRi in the PEAS at DL2. Key secondary endpoints were ORR and rate of undetectable MRD (uMRD; 10 −4 by next-generation sequencing) in blood. All null hypotheses were tested at the primary analysis and not retested in this analysis. Results:Of 137 leukapheresed patients, 118 received liso-cel (safety set), 97 (DL1 = 9; DL2 = 88) were efficacy evaluable, and 54 (DL1 = 4; DL2 = 50) were in the PEAS. In the safety set, median (range) age was 65 years (49-82), 83% had high-risk cytogenetics (del[17p], 42%; TP53 mutation, 47%; unmuted immunoglobulin heavy-chain variable gene, 47%; ≥ 3 chromosomal aberrations, 61%), median (range) lines of prior therapy was 5 (2-14), and all patients had prior BTKi. As of data cutoff (February 28, 2023), median (range) on-study follow-up was 23.5 months (0.4-59.6) for the safety set. In the PEAS at DL2, CR/CRi rate was 20% (95% CI, 10.0-33.7; Table 1). ORR was 44% (95% CI, 30.0-58.7). One patient who had best overall response (BOR) of partial response/remission (PR) at primary analysis had deepened to CR/CRi at 18 months without any additional therapy. Of 9 patients who had BOR of CR/CRi at the primary analysis, 8 had ongoing CR/CRi and 1 completed the study with the last assessment as CR/CRi. The uMRD rate was 64% (95% CI, 49.2-77.1) in blood and 60% (95% CI, 45.2-73.6) in marrow. Median (95% CI) duration of response was 35.3 months (12.4-not reached [NR]) and median duration of CR/CRi was NR. Median (95% CI) PFS was 11.9 months (5.7-26.2). Median (95% CI) OS was 30.3 months (15.0-NR). The efficacy outcomes were similar in the full population at DL2. Of 16 patients who had BOR of CR/CRi at primary analysis, 10 had ongoing CR/CRi. In the safety set, rates of any-grade and grade ≥ 3 treatment-emergent AEs were similar across age groups (Table 2). The rate of any-grade cytokine release syndrome (CRS) was 85% (grade 3, 8%; no grade 4/5) and neurological events (NE) was 45% (grade 3, 18%; grade 4, 1%; no grade 5); 69% received tocilizumab and/or corticosteroids for CRS/NEs. Median (range) time to onset and resolution was 4 days (1-18) and 6 days (2-37) for CRS and 7 days (1-21) and 7 days (1-83) for NEs, respectively. Prolonged cytopenia (grade ≥ 3 at Day 30 after liso-cel infusion), grade ≥ 3 infections, hypogammaglobulinemia, tumor lysis syndrome, second primary malignancy, and macrophage activation syndrome occurred in 54%, 18%, 15%, 11%, 9%, and 3%, respectively. Forty-five (33%) of 137 leukapheresed patients died after liso-cel infusion (disease progression, n = 27 [20%]; AE, n = 6 [4%]; other reasons, n = 12 [9%]). Conclusions: With longer follow-up, liso-cel continued to demonstrate durable CR/CRi, high uMRD rates, and a manageable safety profile in patients with heavily pretreated, high-risk R/R CLL/SLL. The safety results from longer follow-up were similar to those reported in the primary analysis with no new safety signals and were consistent across age groups.

Tyrosine Kinase Inhibitor Withdrawal Syndrome in Chronic Myeloid Leukemia Patients Participants of Two Discontinuation Clinical Trials

Discontinuation of tyrosine kinase inhibitors (TKI) is currently one of the main goals in CML treatment. One of the known adverse events after TKI discontinuation is withdrawal syndrome (WS), which can occur in about 30% of patients (pts), characterized by musculoskeletal pain that begins or worsens weeks after interruption and may be more severe in some cases, but rarely leading to the need for TKI reintroduction. Some studies suggest TKI dose reduction or optimization before stopping treatment may reduce WS rates. Aims: To estimate the incidence of WS after TKI discontinuation in CML in two clinical trials and to correlate the incidence of these events with patients' characteristics and molecular relapse-free survival (MRFS). Methods: We analyzed two cohorts of CML patients of two prospective, single-arm, phase II clinical trials: EDI-PIO (Clinicaltrials.gov: NCT02852486) and DES-CML (ReBEC UTN code : U1111-1252-7312). In both trials, inclusion criteria were CML in the chronic phase, treated with TKI for at least three years, and with deep molecular response (MR4.5) for two years. In the first trial, TKI was discontinued after using three months of pioglitazone, and in the DES-CML trial, after reducing the dose of the TKI by 50% for six months before discontinuation. TKI was reintroduced if the patient lost MR4.0 (EDI-PIO trial) or major molecular response (MMR) (DES-CML trial). Adverse events related to WS were evaluated until the fourth month after discontinuation. WS was defined as the onset of new musculoskeletal pain and/or an increase in pre-existing pain and/or the need to start any pain medication. The adverse event classification followed the Common Terminology Criteria (CTCAE- Version 4.0), musculoskeletal and connective tissue disorder. The data were collected and stored using the REDcap Platform (by Vanderbilt). MRFS was defined by discontinuation date until the last seen date or loss of MR4.0/MMR. The Kaplan-Meier method was applied for OS, whereas Log-Rank tests were applied to compare its curves. All statistical analysis were done by IBM-SPSS v. 24 and considered significant p-value &amp;lt; 5%. Results: Between September 2016 and April 2023, we analyzed 63 patients, 32 from the DES-CML study and 31 from the EDI-PIO trial. The median age at discontinuation was 58 years (24-79), 59% male, 52% Sokal low-risk, 35% intermediate, and 13% high; at discontinuation, 94% of the patients were using Imatinib, 3% Dasatinib, 1.5% Bosutinib and 1.5% Nilotinib. The median duration of TKI treatment until discontinuation was 10 years (3-20). Withdrawal syndrome occurred in 41% of cases and was more frequent in females (58% vs. 42%, P=0.02). Fourteen pts from the EDI-PIO study (45%) and 12 (37.5%) from DES-CML developed WS (P=0.61). According to CTCAE, the events were classified as grade 1, 36%, and 33%, while grade 2 was 57% and 58%; and grade 3 was 7% and 8% (P=0.90), respectively, in EDI-PIO and DES-CML trials. Symptoms started at month 1 in 20% of the patients and 73% in months 2 and 3. Seventeen (30%) pts used anti-inflammatory drugs, analgesics, steroids, or muscle relaxants (64% EDI-PIO vs. 66% DES-CML, P=0.90). There was no need to reintroduce TKI for treating WS. Molecular relapse free-survival (MRFS) at 12 months was 72% (95% CI 60-84%), with no difference by protocol (75% DES-CML, 70% EDI-PIO). MRFS at 12 months was 73% and 74% in the group with or without WS, respectively (P=NS). Conclusions: The incidence and severity of WS were similar in the two trials, corroborating data from the literature. WS was more frequent in females. The dose reduction did not reduce WS rates. However, this condition was self-limited and managed with drugs, without reintroducing TKI. There was no difference in MRFS between the groups with or without WS.

Treatment Patterns and Modifications of Tyrosine Kinase Inhibitors (TKI) Therapy in Early Lines in Patients with Chronic Myeloid Leukemia in Chronic Phase (CML-CP): Real-World Analysis from a Large Commercial Claims Database in the United States (US)

Introduction: Although current TKI options are often successful in managing CML-CP in early lines of therapy, patients may experience issues related to tolerability or resistance to TKIs. Physicians often have to proceed with treatment modifications including discontinuation, switching, temporary interruptions, or dose modifications when the patient's current treatment is not working as expected. This retrospective cohort study assessed these TKI therapy management patterns in patients with CML in first-line (1L) or second-line (2L) treatment in the US. Methods: A large administrative claims database of commercially insured patients, the OptumInsight Clinformatics database (January 2007-June 2022), was used to select adult patients with CML who received 1L TKI therapy (imatinib, dasatinib, nilotinib, or bosutinib; 1L cohort) and those who received 2L TKI therapy (imatinib, dasatinib, nilotinib, bosutinib, or ponatinib; 2L cohort), with treatment initiation (index date) on or after 2012. Patients were required to have ≥6 months of continuous health plan enrollment pre-index, with no hematopoietic stem cell transplantation (HSCT) or CML-related chemotherapy treatments for accelerated phase (AP) or blast crisis (BC) pre-index. Treatment modifications assessed post-index included 1) treatment discontinuation, defined as the earliest of HSCT/treatment switch (defined below)/initiation of a CML-related chemotherapy for AP or BC, 2) treatment switch defined as initiation of another TKI/omacetaxine mepesuccinate, 3) temporary treatment interruption defined as gap of ≥ 14 days followed by resuming the index TKI, and 4) dose reduction (as compared to the initial dose) were assessed. Time to first event post-index were conducted using Kaplan Meier analyses for each of these treatment modifications, in 1L and 2L cohorts, separately. Results: A total of 2,043 patients were included in the 1L cohort with 47.8% of them receiving imatinib, 31.0% dasatinib, 19.2% nilotinib, and 2.0% bosutinib (median age: 64 years; 44.9% female; 69.7% White; 49.5% with Medicare advantage plan operated by commercial insurance [Part C]). Patients in the 1L cohort were observed for a mean (SD) period of 29.1 (27.3) months from index date to earliest of loss of follow-up/end of data. Overall, more than a quarter of patients (27.3%) experienced treatment discontinuation, with rates of discontinuation of 16.1% at 6 months, 25.0% at 12 months, and 33.5% at 24 months post-index ( Figure 1). Treatment switch was observed in 26.0% of the patients, with rates of 15.5% at 6 months and 24.4% at 12 months post-index. The proportion of patients with temporary treatment interruption was 18.6%, with rates of 20.6% at 6 months and 22.2% at 12 months post-index. Dose reduction (as compared to initial dose) occurred in 21.9% of patients, with rates of 18.8% at 6 months and 23.3% at 12 months post-index. A total of 586 patients were included in the 2L cohort with 23.0% of them receiving imatinib, 36.9% dasatinib, 26.6% nilotinib, 10.9% bosutinib and 2.6% ponatinib (median age: 65 years; 47.3% female; 72.5% White; 48.3% with Medicare advantage plan operated by commercial insurance [Part C]). Patients in the 2L cohort were observed for a mean (SD) period of 28.4 (25.4) months from index date to earliest of loss of follow-up/end of data. Almost a third of patients (32.8%) experienced treatment discontinuation, with rates of discontinuation of 22.1% at 6 months, 29.9% at 12 months, and 43.1% at 24 months post-index ( Figure 1). Treatment switch was observed in 30.5% of patients, with rates of 20.5% at 6 months and 28.3% at 12 months post-index. The proportion of patients with temporary treatment interruption was 21.3%, with rates of 24.8% at 6 months and 25.4% at 12 months post-index. Dose reduction as compared to initial dose occurred in 20.6% of patients, with rates of 17.5% at 6 months and 27.2% at 12 months post-index. Conclusions: Findings from this real-world study demonstrate that almost a quarter of patients with CML treated with current TKI treatment options in 1L or 2L experienced treatment discontinuation/switch in the first year from TKI initiation suggesting lack of optimal response/tolerability to currently available TKIs, with more pronounced rates in 2L. This suggests that unmet clinical needs still persist in this population, warranting optimal 1L and 2L treatment options with better tolerability profile for CML management.

SHP2 Inhibition Overcomes Adaptive and Acquired Resistance to FLT3 TKI to Improve Efficacy Against FLT3/ITD AML

FMS-like tyrosine kinase-3 (FLT3) is a receptor tyrosine kinase (RTK) that is frequently mutated in acute myeloid leukemia (AML), with the internal tandem mutation (FLT3/ITD) conferring a poor prognosis to AML patients. This makes FLT3/ITD AML a validated target for treatment with tyrosine kinase inhibitors (TKI), but their success has been hindered by mechanisms of resistance including rapid signal reprogramming known as adaptive resistance and acquired resistance gained through long-term treatment. Past work from our lab has shown that FLT3/ITD cells undergo adaptive resistance, through the rapid reactivation of ERK signaling within 24 hours of sustained FLT3 inhibition. To improve therapy of FLT3/ITD AML it is important to identify therapeutic targets that overcome this ERK reactivation. Since adaptive ERK reactivation is often due to the relief of negative feedback loops that result in the activation of other RTK, we performed an RTK phospho-array on the FLT3/ITD cell line MV4;11. 48-hour treatment with the FLT3 TKI sorafenib resulted in increased activation of IGF-R, HER3 and AXL. While recent work from our lab found that combining AXL and FLT3 inhibition abrogated the ERK reactivation and exerted greater antileukemic effects, the activation of multiple RTKs suggests that targeting a signaling hub that inhibits multiple RTKs at once will likely prove more efficacious in overcoming this adaptive response. Recent literature has implicated SH2-containing protein tyrosine phosphatase 2 (SHP2) as a therapeutic target in adaptive ERK reactivation in other cancers. SHP2 acts directly downstream of RTKs and is involved in the activation of the ERK pathway. Therefore, we explored whether the addition of an SHP2 inhibitorto FLT3 TKI can overcome the adaptive response and increase treatment efficacy. To explore this possibility, we first treated the FLT3/ITD cell lines Molm14 and MV4;11 as well as FLT3/ITD patient samples with FLT3 TKI alone or in combination with an SHP2 inhibitor for 24 hours. The addition of SHP099 to sorafenib decreased the expected ERK reactivation in a dose-dependent manner. Furthermore, combination treatment with synergized to decrease proliferation and increase apoptosis in the FLT3/ITD cell lines. Impressively, in a long-term assay, treatment with the combination for just 72 hours greatly reduced leukemic cell recovery. To determine whether this combination would improve efficacy in vivo, we engrafted NSG mice with Molm14 cells and treated them for 4 weeks via oral gavage. The combination of SHP099 with gilteritinib group resulted in a statistically significantly lower leukemia burden compared to either treatment alone and the vehicle control. The combination group also had prolonged survival compared to the other groups. In addition to rapid adaptive resistance, there are mechanisms of acquired resistance that render traditional FLT3 TKI treatments ineffective against FLT3/ITD AML cells. We next examined the effects of SHP2 inhibition on FLT3-TKI resistant cells models that were generated by growing and maintaining Molm14 and MV4;11 cells in the presence of FLT3 TKI. The combination of FLT3 TKI and SHP099 resensitized the cells to FLT3 TKI, resulting in decreased ERK phosphorylation, decreased proliferation and increased apoptosis of these FLT3 TKI resistant cells. We also explored the effects of treating the FLT3/ITD mutant cell lines with the combination of SHP099 and several of the cytotoxic chemotherapy agents typically used in the treatment of AML including cytarabine and daunorubicin. While daunorubicin or cytarabine alone had little effect on the level of phosphorylated ERK, the addition of SHP099 decreased ERK activation and the combination synergized to exert greater to decrease proliferation and increase apoptosis. These studies reveal that combining FLT3 TKI treatment with an allosteric SHP2 inhibitor decreases the adaptive ERK reactivation observed in FLT3/ITD cells. The finding that SHP099 synergized with both FLT3 TKI and chemotherapy agents in different FLT3-mutated AML models speaks to the versatile efficacy of SHP2 inhibition in multiple AML models. Taken together, the data suggests that SHP2 inhibition can help overcome both adaptive and acquired resistance in FLT3/ITD AML and is a candidate to try to improve patient outcomes.

578P Tyrosine kinase inhibitor treatment of elderly patients with epidermal growth factor receptor mutated advanced non-small cell lung cancer: A multi-institute retrospective study

578P Tyrosine kinase inhibitor treatment of elderly patients with epidermal growth factor receptor mutated advanced non-small cell lung cancer: A multi-institute retrospective study

Evaluating the Potential of Delta Radiomics for Assessing Tyrosine Kinase Inhibitor Treatment Response in Non-Small Cell Lung Cancer Patients.

Our study aimed to harness the power of CT scans, observed over time, in predicting how lung adenocarcinoma patients might respond to a treatment known as EGFR-TKI. Analyzing scans from 322 advanced stage lung cancer patients, we identified distinct image-based patterns. By integrating these patterns with comprehensive clinical information, such as gene mutations and treatment regimens, our predictive capabilities were significantly enhanced. Interestingly, the precision of these predictions, particularly related to radiomics features, diminished when data from various centers were combined, suggesting that the approach requires standardization across facilities. This novel method offers a potential pathway to anticipate disease progression in lung adenocarcinoma patients treated with EGFR-TKI, laying the groundwork for more personalized treatments. To further validate this approach, extensive studies involving a larger cohort are pivotal.

New Generations of Tyrosine Kinase Inhibitors in Treating NSCLC with Oncogene Addiction: Strengths and Limitations.

Tyrosine kinase inhibitors (TKIs) revolutionized the treatment of patients with advanced or metastatic non-small cell lung cancer (NSCLC) harboring most driver gene alterations. Starting from the first generation, research rapidly moved to the development of newer, more selective generations of TKIs, obtaining improved results in terms of disease control and survival. However, the use of novel generations of TKIs is not without limitations. We reviewed the main results obtained, as well as the ongoing clinical trials with TKIs in oncogene-addicted NSCLC, together with the biology underlying their potential strengths and limitations. Across driver gene alterations, novel generations of TKIs allowed delayed resistance, prolonged survival, and improved brain penetration compared to previous generations, although with different toxicity profiles, that generally moved their use from further lines to the front-line treatment. However, the anticipated positioning of novel generation TKIs leads to abolishing the possibility of TKI treatment sequencing and any role of previous generations. In addition, under the selective pressure of such more potent drugs, resistant clones emerge harboring more complex and hard-to-target resistance mechanisms. Deeper knowledge of tumor biology and drug properties will help identify new strategies, including combinatorial treatments, to continue improving results in patients with oncogene-addicted NSCLC.

Osimertinib as a Potential Targeted Therapy for Non-Small Cell Lung Carcinoma (NSCLC) Patients with EGFR Exon 20 T790M

BACKGROUND: Emergence of drug resistance due to epidermal growth factor receptor (EGFR) Exon 20 T790M poses a challenge in the effective management of non-small cell lung carcinoma (NSCLC). Significant breakthrough in the management of NSCLC with a specific genetic alteration causes substantial condition improvement in patients whose cancer progressed after first-generation tyrosine kinase inhibitor treatment and who developed tumors with EGFR Exon 20 T790M mutation. The present study analyzed a cohort of NSCLC patients and investigated the incidence of the EGFR Exon 20 T790M status with Osimertinib therapy, along with its impact on survival rates.METHODS: This was a retrospective cohort study on 22 NSCLC subjects who were genetically examined for EGFR status from plasmic cell free total nucleic acid. Subjects with EGFR Exon 20 T790M mutation were treated with/without Osimertinib. Demographic and clinical data were descriptively summarized, and the differences of each variable and correlation between survival rate and EGFR Exon 20 T790M were analyzed.RESULTS: Subjects with (n=13) and without (n=9) EGFR Exon 20 T790M had survival rates of 10.77±2.45 and 4.78±1.48, respectively (p=0.000). Based on 1-year survival status of subjects with EGFR Exon 20 T790M, there were 3 Osimertinib-treated survivors and 2 Osimertinib-treated non-survivors. Eight subjects with EGFR Exon 20 T790M and without Osimertinib treatment did not survive (p=0.001).CONCLUSION: Since the treatment of Osimertinib demonstrated a noteworthy survival rate among NSCLC subjects with EGFR Exon 20 T790M, thus Osimertinib could be suggested as a potential targeted therapy for NSCLC subjects with EGFR Exon 20 T790M.KEYWORDS: non-small cell lung carcinoma, EGFR, Exon 20, T790M, osimertinib

Identification of multivariable microRNA and clinical biomarker panels to predict imatinib response in chronic myeloid leukemia at diagnosis.

Imatinib Mesylate (imatinib) was once hailed as the magic bullet for chronic myeloid leukemia (CML) and remains a front-line therapy for CML to this day alongside other tyrosine kinase inhibitors (TKIs). However, TKI treatments are rarely curative and patients are often required to receive life-long treatment or otherwise risk relapse. Thus, there is a growing interest in identifying biomarkers in patients which can predict TKI response upon diagnosis. In this study, we analyze clinical data and differentially expressed miRNAs in CD34+ CML cells from 80 patients at diagnosis who were later classified as imatinib-responders or imatinib-nonresponders. A Cox Proportional Hazard (CoxPH) analysis identified 16 miRNAs that were associated with imatinib nonresponse and differentially expressed in these patients. We also trained a machine learning model with different combinations of the 16 miRNAs with and without clinical parameters and identified a panel with high predictive performance based on area-under-curve values of receiver-operating-characteristic and precision-recall curves. Interestingly, the multivariable panel consisting of both miRNAs and clinical features performed better than either miRNA or clinical panels alone. Thus, our findings may inform future studies on predictive biomarkers and serve as a tool to develop more optimized treatment plans for CML patients in the clinic.

Triple combination of HAIC-FO plus tyrosine kinase inhibitors and immune checkpoint inhibitors for advanced hepatocellular carcinoma: A systematic review and meta-analysis.

The triple combination of hepatic arterial infusion chemotherapy (HAIC) with fluorouracil, leucovorin, and oxaliplatin (FOLFOX) plus tyrosine kinase inhibitor (TKI) and immune checkpoint inhibitors (ICIs) is expected to have a synergistic anticancer effect in HCC. We conducted this meta-analysis to evaluate the efficacy and safety of the triple combination treatment in advanced HCC patients. PubMed, Embase, Cochrane Library, Web of Science databases were systematically searched for relevant studies from the inception of each database to May 10, 2023. All articles focusing the triple combination treatment of HAIC-FO plus TKI and ICIs for advanced HCC were eligible. The meta-analysis was conducted following the PRISMA guidelines. The risk of bias was assessed using the Joanna Briggs Institute (JBI) for case series and Newcastle-Ottawa Scale (NOS) for cohort studies. The primary outcomes were overall survival (OS), progression-free survival (PFS), objective response rate (ORR) and disease control rate (DCR). The secondary results were adverse events. Further meta-analysis of control studies demonstrated the superiority of the triple combination modality to TKI plus ICIs, and TKI alone. Nine articles (four cohort studies and five one-arm studies) involving 777 advanced HCC patients were included in this meta-analysis. In terms of survival analysis, the pooled median PFS was 11 months (95% CI: 10.1-12.0 months) with low heterogeneity (I2 = 0%, p = 0.97). With regard to tumor response, the pooled ORR and DCR was 61.6% (I2=0%, p = 0.71) and 87.9% (I2 = 13%, p = 0.33) with low heterogeneity, respectively. As compared with TKIs plus ICIs, and TKIs alone, the triple combination thrapy was associated with improved median OS (HR=0.51, 95%CI 0.41-0.62) with low heterogeneity across studies (I2 = 0%, p = 0.47), median PFS (HR=0.51, 95%CI 0.41-0.64) with low heterogeneity across studies (I2 = 0%, p = 0.41), ORR (RR = 0.56, 95% CI: 0.42-0.74) with high heterogeneity across studies (I2 = 69%, p = 0.02), and DCR (RR = 0.38, 95%CI 0.27-0.54) with low heterogeneity across studies (I2 = 14%, p = 0.32). The most common 3/4 AEs were elevated ALT and AST, thrombocytopenia, hypertension, nausea and vomiting in this meta-analysis. The triple combination therapy of HAIC-FO plus TKI and ICIs showed promising efficacy and safety in patients with advanced HCC. The protocol was registered with PROSPERO (ID:CRD42023424281).

Co-Occurring Alterations in Multiple Tumor Suppressor Genes Are Associated With Worse Outcomes in Patients With EGFR-Mutant Lung Cancer

IntroductionPatients with metastatic EGFR-mutant NSCLC inevitably have disease progression while on tyrosine kinase inhibitor (TKI) therapy. Co-occurring tumor suppressor gene (TSG) alterations have been associated with poor outcomes, however, detailed analyses of their impact on patient outcomes are limited. MethodsPatients with EGFR-mutant NSCLC treated with EGFR TKIs who had tumor genomic profiling were included. Alterations in TP53 and five additional TSGs (RB1, NF1, ARID1A, BRCA1, and PTEN) were used to stratify the cohort into the following three subgroups: patients with tumors harboring a TP53 mutation plus a mutation in at least one additional TSG (TP53mut/TSGmut), those having a TP53 mutation without additional TSG mutations (TP53mut/TSGwt), and those with TP53wt. Patient characteristics and clinical outcomes were assessed in two independent cohorts. ResultsA total of 101 patients from the Yale Cancer Center and 182 patients from the American Association for Cancer Research Project GENIE database were included. In the Yale cohort, TP53 mutations were identified in 65 cases (64%), of which 23 were TP53mut/TSGmut and 42 were TP53mut/TSGwt. Although the presence of a TP53 mutation was associated with worse outcomes, the additional TSG alteration in TP53mut tumors identified a subset of patients associated with particularly aggressive disease and inferior clinical outcome in both the Yale and the GENIE cohorts. Specifically, in the Yale cohort for patients receiving first-line TKIs, those with TP53mut/TSGmut tumors had shorter progression-free survival (PFS) and overall survival (OS) than TP53mut/TSGwt (PFS: hazard ratio [HR] = 2.03, confidence interval [CI]: 1.12–3.69, p < 0.01, OS: HR = 1.58, CI: 0.82–3.04, p = 0.12) or TP53wt cases (PFS: HR 2.4, CI: 1.28–4.47, p < 0.001, OS: HR = 2.54, CI: 1.21–5.34, p < 0.005). Inferior outcomes in patients with TP53mut/TSGmut tumors were also found in those receiving osimertinib as second-line therapy. Similar findings were seen in patients in the GENIE cohort. ConclusionsPatients with TP53mut/TSGmut tumors represent a patient subgroup characterized by an aggressive disease phenotype and inferior outcomes on EGFR TKIs. This information is important for understanding the biological underpinnings of differential outcomes with TKI treatment and has implications for identifying patients who may benefit from additional therapeutic interventions beyond osimertinib monotherapy.

MIG6 Mediates Adaptive and Acquired Resistance to ALK/ROS1 Fusion Kinase Inhibition through EGFR Bypass Signaling.

Despite the initial benefit from tyrosine kinase inhibitors (TKI) targeting oncogenic ALK and ROS1 gene fusions in non-small cell lung cancer, complete responses are rare and resistance ultimately emerges from residual tumor cells. Although several acquired resistance mechanisms have been reported at the time of disease progression, adaptative resistance mechanisms that contribute to residual diseases before the outgrowth of tumor cells with acquired resistance are less clear. For the patients who have progressed after TKI treatments, but do not demonstrate ALK/ROS1 kinase mutations, there is a lack of biomarkers to guide effective treatments. Herein, we found that phosphorylation of MIG6, encoded by the ERRFI1 gene, was downregulated by ALK/ROS1 inhibitors as were mRNA levels, thus potentiating EGFR activity to support cell survival as an adaptive resistance mechanism. MIG6 downregulation was sustained following chronic exposure to ALK/ROS1 inhibitors to support the establishment of acquired resistance. A higher ratio of EGFR to MIG6 expression was found in ALK TKI-treated and ALK TKI-resistant tumors and correlated with the poor responsiveness to ALK/ROS1 inhibition in patient-derived cell lines. Furthermore, we identified and validated a MIG6 EGFR-binding domain truncation mutation in mediating resistance to ROS1 inhibitors but sensitivity to EGFR inhibitors. A MIG6 deletion was also found in a patient after progressing to ROS1 inhibition. Collectively, this study identifies MIG6 as a novel regulator for EGFR-mediated adaptive and acquired resistance to ALK/ROS1 inhibitors and suggests EGFR to MIG6 ratios and MIG6-damaging alterations as biomarkers to predict responsiveness to ALK/ROS1 and EGFR inhibitors.