20457634, 2021, 17, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4135 by Readcube (Labtiva Inc.), Wiley Online Library on [16/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Received: 13 May 2021 | Revised: 30 June 2021 | Accepted: 1 July 2021 DOI: 10.1002/cam4.4135 RESEARCH ARTICLE LungBEAM: A prospective multicenter study to monitor stage IV NSCLC patients with EGFR mutations using BEAMing technology Pilar Garrido1,2 | Luis Paz-­Ares2,3,4,5 | Margarita Majem6,7 | Teresa Morán7,8 | José Manuel Trigo9 | Joaquim Bosch-­Barrera10 | Rosario Garcίa-­Campelo11 | José Luis González-­Larriba12 | José Miguel Sánchez-­Torres13 | Dolores Isla14 | Núria Viñolas15 | Carlos Camps2,16 | Amelia Insa17 | Óscar Juan18 | Bartomeu Massuti19 | Alfredo Paredes20 | Ángel Artal21 | Marta López-­Brea22 | José Palacios2,23 | Enriqueta Felip24 1 Medical Oncology Department, IRYCIS Hospital Universitario Ramón y Cajal, Universidad Alcalá, Madrid, Spain 2 CIBERONC, Madrid, Spain 3 Medical Oncology Department, Hospital Universitario 12 de Octubre and i+12 Research Institute, Madrid, Spain 4 Lung Cancer Group, Clinical Research Program, Spanish National Cancer Research Center (CNIO), Madrid, Spain 5 Complutense University, Madrid, Spain 6 Medical Oncology Department, Hospital De La Santa Creu I Sant Pau, Barcelona, Spain 7 Spanish Lung Cancer Group (GECP), Barcelona, Spain 8 ICO Badalona, Hospital Germans Trias i Pujol, Barcelona, Spain 9 Medical Oncology Department, Hospital Universitario Virgen de la Victoria, Málaga, Spain 10 Medical Oncology, Catalan Institute of Oncology (ICO), Dr. Josep Trueta Hospital of Girona, Girona, Spain 11 Medical Oncology Department, Hospital Universitario Da Coruña, A Coruña, Spain 12 Medical Oncology Department, Hospital Universitario Clínico San Carlos, Madrid, Spain 13 Medical Oncology Department, Hospital de la Princesa, Madrid, Spain 14 Medical Oncology Department, Hospital Universitario Lozano Blesa, Zaragoza, Spain 15 Medical Oncology Department, Hospital Clinic i Provincial, Barcelona, Spain 16 Medical Oncology Department, Hospital General Universitario de Valencia, Universidad de Valencia, Valencia, Spain 17 Medical Oncology Department, Hospital Clínico Universitario, Valencia, Spain 18 Medical Oncology Department, Hospital Universitario La Fe, Valencia, Spain 19 Medical Oncology Department, Hospital General Universitario de Alicante, Alicante, Spain 20 Medical Oncology Department, Hospital Universitario Donostia, Donostia-­San Sebastián, Spain 21 Medical Oncology, Hospital Universitario Miguel Servet, Zaragoza, Spain 22 Medical Oncology Department, Hospital Universitario Marqués de Valdecilla, Santander, Spain 23 Pathology Department, IRYCIS Hospital Universitario Ramón y Cajal, Universidad Alcalá, Madrid, Spain 24 Medical Oncology Department, Hospital Universitario Vall d’Hebron, Barcelona, Spain This is an open access article under the terms of the Creat​ive Commo​ns Attri​bution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2021 The Authors. Cancer Medicine published by John Wiley & Sons Ltd. 5878 |  wileyonlinelibrary.com/journal/cam4 Cancer Medicine. 2021;10:5878–5888. | 20457634, 2021, 17, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4135 by Readcube (Labtiva Inc.), Wiley Online Library on [16/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GARRIDO et al.    5879 Correspondence Pilar Garrido, Head of Thoracic Tumor Abstract Unit, Medical Oncology Department, Objectives: The aim of LungBEAM was to determine the value of a novel epidermal Hospital Universitario Ramón y Cajal, growth factor receptor (EGFR) mutation test in blood based on BEAMing technol- Carretera Colmenar Viejo KM 9100, 28034 Madrid, Spain. ogy to predict disease progression in advanced non-­small cell lung cancer (NSCLC) Email: pilargarridol@gmail.com patients treated with first-­or second-­generation EGFR-­tyrosine kinase inhibitors Funding information (EGFR-­TKIs). Another goal was to monitor the dynamics of EGFR mutations, as This work was supported by Sysmex well as to track EGFR exon 20 p.T790M (p.T790M) resistance during treatment, as Inostics GmbH. The sponsor and the critical indicators of therapeutic efficacy and patient survival. study coordinating investigators were involved in the study design and data Methods: Stage IV NSCLC patients with locally confirmed EGFR-­ TKI sensi- interpretation. Writing and editorial tizing mutations (ex19del and/or L858R) in biopsy tissue who were candidates to assistance was funded by Sysmex Inostics receive first-­or second-­generation EGFR-­TKI as first-­line therapy were included. GmbH. Plasma samples were obtained at baseline and every 4 weeks during treatment until a progression-­free survival (PFS) event or until study completion (72-­week follow-­up). The mutant allele fraction (MAF) was determined for each identified mutation using BEAMing. Results: A total of 68 of the 110 (61.8%) patients experienced a PFS event. Twenty-­ six patients (23.6%) presented with an emergent p.T790M mutation in plasma at some point during follow-­up, preceding radiologic progression with a median of 76 (inter- quartile ratio: 54–­111) days. Disease progression correlated with the appearance of p.T790M in plasma with a hazard ratio (HR) of 1.94 (95% confidence interval [CI], 1.48–­2.54; p < 0.001). The HR for progression in patients showing increasing plasma sensitizing mutation levels (positive MAF slope) versus patients showing either de- creasing or unchanged plasma mutation levels (negative or null MAF slopes) was 3.85 (95% CI, 2.01–­7.36; p < 0.001). Conclusion: Detection and quantification of EGFR mutations in circulating tumor DNA using the highly sensitive BEAMing method should greatly assist in optimizing treatment decisions for advanced NSCLC patients. KEYWORDS BEAMing, EGFR mutations, liquid biopsy, non-­small cell lung carcinoma 1 | IN T RO D U C T IO N detection is, therefore, a minimally invasive method for therapy response monitoring in NSCLC patients.3,4 Finally, Epidermal growth factor receptor (EGFR) genotyping is a EGFR mutation results detected via liquid biopsy can pre- routine test to assess non-­small-­cell lung cancer (NSCLC) cede radiologic progression by weeks, and in some cases, patients’ eligibility to receive EGFR tyrosine kinase inhibitor even months.5 (TKI) therapy. An additional EGFR mutation, known as exon As of the present day, few prospective trials have com- 20 p.T790 M (p.T790 M), has been shown to be the most prehensively monitored both sensitizing and resistance muta- common cause of resistance to first-­and second-­generation tions in plasma samples during first-­and second-­generation EGFR-­TKIs; it has been observed in 50%–­60% of cases of ac- TKI treatment.6-­9 With respect to EGFR p.T790 M, for in- quired resistance.1 A meta-­analysis has shown that p.T790 M stance, it has been demonstrated that the risk of disease mutation emergence during EGFR-­TKI therapy has a nega- progression during therapy with first-­or second-­generation tive impact on progression-­free survival (PFS),2 and since EGFR-­TKIs is known to increase when the mutant allele tumor tissue biopsy or re-­biopsy is not always viable or ef- fraction (MAF) of a preexisting p.T790 M mutation attains a fective, analysis of circulating tumor-­derived DNA (ctDNA; threshold of ≥3.2%.10 Therefore, ascertaining MAF from the also known as liquid biopsy) is a feasible alternative to tissue-­ analyses of plasma-­detected EGFR sensitizing mutations is based EGFR mutation testing. Plasma-­based EGFR mutation likely to improve gauging the prognosis of patients receiving | 20457634, 2021, 17, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4135 by Readcube (Labtiva Inc.), Wiley Online Library on [16/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 5880    GARRIDO et al. EGFR-­TKIs. For example, a reduced benefit of EGFR-­TKI principles of the Declaration of Helsinki. All eligible patients therapy has been associated with increasing levels of sensi- signed the informed consent form before any study-­specific tizing mutations in ctDNA, likely reflecting an increase in procedures, sampling, or analyses. systemic tumor burden and/or overall tumor activity.9 These findings underscore the importance of detecting and quanti- fying EGFR mutations in NSCLC patients before and during 2.2 | Study procedures TKI therapy with the aim of optimizing the therapeutic regimen.11 Plasma samples were obtained from 10 ml of blood collected Although third-­generation TKIs such as osimertinib are prior to any therapeutic intervention and every 4 weeks dur- approved for use in the first-­line setting, their adoption into ing treatment until disease progression, death, or completion routine clinical practice varies widely.12 Furthermore, first-­ of the study follow-­up (72 weeks after treatment initiation). and second-­generation EGFR-­TKIs are prominently used in After collection, plasma samples were frozen and then sent several cancer care settings—­therefore, it remains a matter to either the Sysmex Inostics GmbH laboratory in Hamburg of urgency to be able to accurately and sensitively recognize (Germany) or the Sysmex Inostics Inc CLIA laboratory in the presence of p.T790 M in NSCLC patients treated with Baltimore (MD, USA) for EGFR mutation testing using the these targeted therapies, at the earliest possible time point. OncoBEAM® EGFR assay. The OncoBEAM® EGFR assay, By detecting p.T790 M mutation as the most likely harbin- based on BEAMing technology, detects eight EGFR muta- ger of treatment resistance that drives disease progression, tions (p.T790 M, L858R, and six common del19 variants). an informed decision can be made for the patient to switch The MAF was recorded. Tissue-­ based mutation profiles to osimertinib or an alternative therapy immediately upon were determined using the standard-­of-­care reference meth- the radiologic confirmation of disease progression. A highly odology used routinely in each hospital laboratory. For any sensitive liquid biopsy platform ideally suited for p.T790 M discordant cases, BEAMing was also performed on tissue resistance detection is the digital polymerase chain reaction samples, when possible, to provide adjudication of results. (PCR)-­based assay BEAMing (beads, emulsions, amplifica- Disease progression was assessed according to the tion, and magnetics). This assay reliably achieves a low limit Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 of detection (0.02% MAF)13 and has been widely utilized in criteria. Computed tomography (CT) scans were performed clinical settings.14-­19 every 8 weeks within the first 6 months and subsequently The main goal of the LungBEAM study was to provide a every 12 weeks (Table S1), unless any clinical condition pragmatic and real-­world approach to evaluate the value of required more frequent imaging. Investigators were blinded EGFR mutation testing, using BEAMing technology, to mon- to plasma BEAMing results. All treatments and diagnostic itor resistance and sensitizing mutations in plasma ctDNA procedures were performed according to the standard clinical and predict disease progression. An underlying imperative practice in each center. was to examine the correlation between the levels of EGFR sensitizing and p.T790 M mutations in plasma during the course of therapy and patient disease trajectory and survival 2.3 | Statistical analysis after EGFR-­TKI treatment. A total of 100 patients (72 PFS events expected at 18-­ month follow-­up) were estimated to be included in the study 2 | M ATE R IA L S A N D ME T HODS to achieve 80% statistical power to detect the effect of the p.T790 M mutation at the 95% confidence level. We assumed 2.1 | Study design and patients that (i) the maximum time to progression is about 16 months in patients with an EGFR sensitizing mutation who receive LungBEAM was a multicenter, prospective study that used first-­or second-­generation EGFR-­TKI therapy1; (ii) about BEAMing technology for mutation analysis in plasma ctDNA 50% of patients will develop a p.T790 M mutation1,20; and in stage IV NSCLC patients. Recruitment was conducted (iii) the hazard ratio (HR) for PFS associated with the inci- in 19 Spanish centers from November 2015 to May 2017. dence of a p.T790 M mutation is around 2.2 Eligible patients included adults with stage IV NSCLC, with Categorical variables are presented using absolute and a locally confirmed EGFR-­TKI sensitizing mutation (ex19del relative frequencies, and continuous variables by dispersion and/or L858R) by tissue biopsy, eligible to receive EGFR-­ measures, mainly median and interquartile range (IQR). PFS TKI therapy, and without prior treatment with EGFR-­TKIs was defined as the time elapsed from the date of treatment or chemotherapy. Patients having concomitant malignancies initiation to disease progression or death, or, alternatively, were excluded. The study was approved by the institutional to the date of the last blood sample collected within the 72-­ review board of each site and was conducted according to the week follow-­up period, whichever came first. A PFS event | 20457634, 2021, 17, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4135 by Readcube (Labtiva Inc.), Wiley Online Library on [16/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GARRIDO et al.    5881 T A B L E 1 Summary of baseline patient and tumor characteristics was defined as disease progression or death. Comparisons of and mutational analyses clinical variables in patients with or without p.T790 M mu- tation were performed using the Chi-­square (χ2) test, t-­test, All or Mann–­Whitney U test. Univariate and multivariate Cox patients Variable, n (%) (N=110) models adjusted to clinical risk factors were used to investi- gate the effect of EGFR mutations detected in plasma on the Age, yearsa 65.5 (12.5) PFS. In all models, the proportional hazard assumption was Gender tested using Schoenfeld residuals. Male 31 (28.2) Agreement between plasma and tissue EGFR mutation Female 79 (71.9) results was determined by Cohen's Kappa statistic. The Smoking status slope of post-­baseline MAF of EGFR sensitizing mutations, Smoker 9 (8.2) defined as the mean change in MAF value for each patient, Ex-­smoker (1–­5 years ago) 11 (10.0) was calculated to characterize patient subgroups according Ex-­smoker (6–­10 years ago) 3 (2.7) to the differential patterns of progression in plasma mutation levels (measured in MAF), which were compared using the Ex-­smoker (>10 years ago) 19 (17.3) Wilcoxon Rank Sum test. Positive and negative slopes in- Never 68 (61.8) dicate increasing and decreasing MAF values, respectively, ECOG over time. Time-­to-­progression according to MAF patterns 0 50 (45.5) was analyzed using the Kaplan–­Meier method and log-­rank 1 50 (45.5) testing. 2 8 (7.3) All statistical tests were performed with two-­sided 95% Not available 2 (1.8) confidence interval (CI) and 5% significance level. Statistical Stage M (at primary tumor diagnosis) analyses were performed using the R 3.5.2 statistical software. M0 13 (11.8) M1 97 (88.2) 3 | R E S U LTS Metastasis location M1a 28 (25.5) 3.1 | Study population M1b 78 (74.5) Number of metastatic locations A total of 110 patients with stage IV EGFR mutated NSCLC 1 56 (50.9) met the selection criteria and were included in the study. The ≥2 54 (49.1) baseline characteristics of the cohort are shown in Table 1. Tissue biopsy The majority of patients were women (71.8%), non-­smokers Primary tumor 85 (77.3) (61.8%), and presented with synchronous diagnosis of pri- Metastasis 25 (22.7) mary tumor and metastatic disease (88.2%). A total of 22.7% of patients had brain metastasis at baseline. Tissue EGFR EGFR mutation (tissue) mutation distribution was 64.5% ex19del and 35.5% L858R; Ex19del 71 (64.5) one patient (0.9%) harbored a tissue p.T790 M mutation at L858R 39 (35.5) baseline. p.T790 M mutation (tissue) During the 72-­ week follow-­ up period, a total of 68 Present 1 (0.9) (61.8%) patients experienced a PFS event (Table 1; Figure 1). Absent 51 (46.4) Follow-­up was lost or incomplete in seven cases. Overall, the Not evaluated 58 (52.7) median (IQR) follow-­up was 352 (164–­504) days. Progression type Extrapulmonary 22 (38.6) 3.2 | Concordance between plasma Pulmonary 17 (29.8) BEAMing results with those from tissue testing Both 14 (24.6) Not available 4 (7.0) Baseline EGFR sensitizing mutations were detected in the Abbreviations: ECOG, Eastern Cooperative Oncology Group; SD, standard plasma of 70.9% (78/110) of the patients. Of these baseline deviation. plasma sensitizing mutation-­positive patients, 67.9% (53/78) a Data are presented as mean (SD). presented with ex19del, while 32.1% (25/78) had L858R mutations. The overall sensitivity for detection of plasma sensitizing mutations by BEAMing was 70.9% (95% CI, | 20457634, 2021, 17, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4135 by Readcube (Labtiva Inc.), Wiley Online Library on [16/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 5882    GARRIDO et al. F I G U R E 1 Study flow-­chart. EGFR, epidermal growth factor receptor; NSCLC, non-­small cells lung cancer; PD, progressive disease; RECIST, Response Evaluation Criteria In Solid Tumors F I G U R E 2 MAF values profile in p.T790 M-­positive patients with progression disease. The event day was taken as a reference to better clarify the evolution of MAF values toward the event time point. One patient was removed to scale the graph because of its high values (up to 12%). MAF, mutant allele fraction 61.4–­79.0). In patients with baseline sensitizing mutations of p.T790 M at baseline was 100%, with a Kappa index of detected in plasma, the concordance of plasma results with 100% (95% CI, 100–­100). those obtained from tissue testing was 98.7% (Kappa: 97%; 95% CI, 91.2–­102.8; Table S2). Among the 32 patients showing different tissue and 3.3 | Detection of p.T790 M mutation in plasma EGFR mutation results (positive tissue and negative plasma and risk of disease progression plasma), additional BEAMing tests were performed on 15 available initial specimens used for standard-­of-­care tissue Twenty-­six patients (23.6%) had a p.T790 M mutation de- testing, and valid data were obtained in 14 cases. The con- tected in their plasma at some point during follow-­up. Of cordance of the EGFR mutations detected by the BEAMing these, 19 (73.1%) had a PFS event during the study period. technique in tissue with the standard-­of-­care tissue result was Of all enrolled patients who progressed (68 patients), 19 100% (95% CI, 71.1–­100; Table S2). (27.9%) presented with a p.T790 M mutation in their plasma The presence or absence of a p.T790 M resistance mu- during the 72-­week follow-­up. There were no significant tation in tissue at baseline was only able to be assessed in differences in demographic or clinical-­pathologic factors 52 patients—­this was due to the limitations of the reference between plasma p.T790 M-­positive and p.T790 M-­negative method in some centers. The sensitivity of BEAMing to detect patients (Table S3). The median of maximum MAF values p.T790 M in plasma was 100% (95% CI, 100–­100; Table S2). in p.T790 M-­positive patients was 0.62% (IQR: 0.15–­1.88). The concordance between tissue and plasma for the detection Significant differences in plasma EGFR sensitizing mutation | 20457634, 2021, 17, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4135 by Readcube (Labtiva Inc.), Wiley Online Library on [16/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GARRIDO et al.    5883 MAF levels were observed between patients with or without (Figure S2). The median baseline MAF detected in these pa- p.T790 M (0.20 vs. 2.50; p < 0.001), regardless of the initial tients was 0.79% (IQR: 0.22–­7.56). Among these patients, baseline level of sensitizing mutations (Table S4). 53 (67.9%) had a PFS event during the 72-­week follow-­up The finding of a p.T790 M resistance mutation in plasma period (Table S8). No significant differences in the distribu- preceded the corresponding PFS event by a median of 76 (IQR: tion of sensitizing mutations with respect to both mutation 54–­111) days. In seven (36.8%) of these patients, the p.T790 M type and exon location were observed between patients with mutation was detected within the first 16 weeks of follow-­up or without PFS events. (Table S5). Of note, one patient (ID_10) presented with, and Among the 78 patients with plasma baseline EGFR sen- persistently showed, high p.T790 M levels (42.9% MAF at base- sitizing mutations, 68 of them had at least four post-­baseline line); this patient was confirmed to harbor a germline p.T790 M samples available and were used to assess the longitudinal mutation and did not experience disease progression at any point patterns of EGFR mutation MAF levels. A marked reduc- during the 72-­week follow-­up (Figure S1; Table S5). Plasma tion of MAF values from baseline levels was observed early monitoring revealed an overall increase in p.T790 M MAF val- during treatment in all patients (Figure 3A), dropping to zero ues up until radiologic progression (Figure 2; Table S6). in some cases. After this initial precipitous drop in MAF, The HR for disease progression associated with detection most patients with a PFS event showed a rebound, with MAF of p.T790 M resistance mutations in plasma was 1.94 (95% values increasing progressively at a variable rate. In contrast, CI, 1.48–­2.54; p < 0.001). This effect remained significant most PFS event-­free patients showed relatively few changes even after the addition of other risk factors as confounders in MAF levels, although some random variations were ob- (e.g., gender, smoking or multiple metastases, HR: 2.08; 95% served (Figure 3B). Patients with PFS events tended to have CI, 1.58–­2.74; Table S7). After incorporation of plasma sen- significantly higher MAF slope values than event-­free pa- sitizing mutations as predictors, the effect of p.T790 M was tients (Wilcoxon Rank Sum Test p < 0.001). Most patients decreased but remained significant (HR: 1.38; 95% CI, 1.02–­ with PFS events (65.1%; 28/43) showed positive MAF slope 1.87; p < 0.038; Table 2). Detection of sensitizing mutations in values, whereas most event-­free patients had negative or null plasma by BEAMing was significantly associated with disease slopes (84.0%; 21/25). progression when adjusted for other risk factors (Table 2). There Progression-­free survival was shorter in patients showing was a higher impact of the mutation on disease progression with a positive MAF slope than in those with a negative or null mutation Ex19del in exon 19, where the HR for disease progres- slope (Figure 4). The median PFS was 296.5 (IQR: 248–­393) sion was 2.12 (95% CI 1.67–­2.70), than with mutation L858R in days in patients with a positive slope, but non-­estimable in exon 21 (HR 1.68; 95% CI 1.27–­2.22, both p < 0.001). patients with a negative or null slope. The Kaplan–­Meier es- timates of 1-­year PFS rates were 40.6% (95% CI, 26.7–­61.8) and 74.3% (95% CI, 61.2–­90.3) for patients with positive and 3.4 | Detection of sensitizing mutations in negative/null slopes, respectively. plasma and risk of disease progression The HR for disease progression in patients with a posi- tive MAF slope was 3.85 (95% CI, 2.01–­7.36; p < 0.001). As stated above, baseline plasma sensitizing mutations were The effect remained significant when adding exon type as a detected in 70.9% (78/110) of patients enrolled in the study confounder (HR: 4.11; 95% CI, 2.11–­8.03; p < 0.001). No effect of exon type on PFS was observed (HR: 1.10; 95% CI, T A B L E 2 Risk factors of progression during the 72-­week follow-­ 0.57–­2.13; p = 0.765). In the multivariate model, there was up in multivariate Cox regression analysis no association between mutated EGFR exon type and PFS (HR: 0.76; 95% CI, 0.39–­1.49; p = 0.423). Variable HR (95% CI) p value Age 1.02 (1.01–­1.03) <0.001 Man 1.06 (0.87–­1.29) 0.584 4 | DISCUSSION Smoker 1.42 (1.03–­1.94) 0.030 Ex-­smoker (1–­5 years ago) 1.53 (1.13–­2.06) 0.006 Precision medicine in the setting of NSCLC is now routine Ex-­smoker (6–­10 years ago) 0.77 (0.43–­1.38) 0.379 due to molecular profiling strategies, which can identify gene Ex-­smoker (>10 years ago) 1.00 (0.78–­1.28) 0.992 alterations in a patient's tumor that are targeted by effective Metastasis (>1 location) 1.27 (1.06–­1.53) 0.011 therapies. Due to the drawbacks of tissue biopsy (and re-­ biopsy) for identifying and monitoring EGFR mutations, the p.T790 M Mutation 1.38 (1.02–­1.87) 0.038 liquid biopsy approach is a viable, non-­invasive, rapid, and Ex19del 2.12 (1.67–­2.70) <0.001 accurate alternative for detecting both EGFR sensitizing and L858R 1.68 (1.27–­2.22) <0.001 resistance mutations. Plasma EGFR mutation testing with Abbreviations: CI, confidence interval; HR, hazard ratio. rapid turnaround time enables physicians to quickly identify | 20457634, 2021, 17, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4135 by Readcube (Labtiva Inc.), Wiley Online Library on [16/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 5884    GARRIDO et al. F I G U R E 3 Progression of MAF values stratified by sensitizing mutation exon type and patient outcome (PFS event). A square root scale of MAF values was used to better visualize MAF progression in patients with very low values. A, Baseline and 4 weekly follow-­up MAF values. B, Baseline values are excluded to clarify MAF patterns. MAF, mutant allele fraction; PFS, progression-­free survival initiation of EGFR-­TKI therapy) compared with these other studies.1,22,23 The emergence of the EGFR p.T790 M mutation during EGFR-­TKI treatment was associated with a shorter PFS, and this was independent of other clinical factors. This result is in contrast to those of previous reports indicating that per- sistence of p.T790 M expression represents indolent progres- sion.20 Notably, our results showed that p.T790 M detection in plasma occurs at a median of 76 days prior to radiologi- cal disease progression. In addition, there was a distinct in- crease in p.T790 M MAF values prior to or at the time of F I G U R E 4 PFS according to the slope (mean increase) of MAF disease progression in p.T790 M-­positive patients. In one-­ in patients with a confirmed EGFR sensitizing mutation in plasma at baseline. EGFR, epidermal growth factor receptor; MAF, mutant allele third of these patients, the p.T790 M mutation was detected fraction; PFS, progression-­free survival during the first 16 weeks of TKI treatment. Early detection of p.T790 M mutation in plasma is important because it helps identify patients that may benefit from earlier and more fre- those patients with EGFR mutation-­positive disease, enabling quent imaging to monitor disease progression, which can bet- early and optimal treatment of these patients that ultimately ter inform subsequent therapeutic interventions without the improves outcomes.21 In this real-­world LungBEAM study, need for an invasive tissue biopsy procedure.24,25 In addition, almost 62% of stage IV NSCLC patients with EGFR sen- the evidence shown in the present study indicates that the tra- sitizing mutations that received first-­and second-­generation jectory of increasing EGFR mutation levels (MAF) is an im- TKI treatment as first-­line therapy had a PFS event dur- portant signal that a patient's tumor burden is increasing.26,27 ing the 72-­week follow-­up period, with 27.9% of patients This principle has also been observed during the monitoring showing an emergent plasma p.T790 M mutation detected of colorectal cancer patients.28,29 Additional studies will be using BEAMing at some points during the follow-­up. This required to fully explore the prognostic impact of increasing percentage of patients was lower than reported in previous plasma EGFR mutation levels in NSCLC. studies; however, this could be attributed to the considerably At the initiation of this study, third-­generation EGFR-­ shorter duration of follow-­up in our study (72 weeks from the TKIs such as osimertinib were under experimental evaluation | 20457634, 2021, 17, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4135 by Readcube (Labtiva Inc.), Wiley Online Library on [16/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GARRIDO et al.    5885 as a potential second-­line treatment after first-­and second-­ (mean change) is a measure of the rate of MAF increase ob- generation EGFR-­TKIs such as erlotinib. Hence, the initial served after the initial drop. These slopes allowed us to clas- objective was to monitor the early appearance of EGFR sify patients into two groups: those who had a progressive p.T790 M mutation in plasma, a well-­established mechanism increase in MAF values (positive slope) and those who did of resistance to first-­and second-­generation TKIs. Although not (null or negative slopes). Comparison of these two groups the clinical application of plasma p.T790 M detection and of patients showed that they experienced clinically meaning- monitoring has fallen as a result of the increase in the use ful differences in 1-­year PFS rates: these rates were signifi- of third-­generation TKIs as first-­line treatment,30 it is none- cantly lower in those with a positive slope versus those with theless still vital to detect p.T790 M mutations in the var- a null or negative slope (40.6% vs. 74.2%). Thus, monitoring ious treatment settings where first-­and second-­generation plasma EGFR mutations and MAF levels using BEAMing TKIs remain in use.31 Moreover, sequential use of second-­ during the follow-­up period enabled accurate measurement of generation EGFR-­TKIs followed by osimertinib might be one a patient's response to treatment. In line with this idea, it has of the most practical ways of improving overall survival in been proposed that sensitizing mutations, such as ex19del, patients that develop p.T790 M resistance. There are a num- can accurately reflect tumor biology, and it could be used as ber of current clinical trials exploring sequential EGFR-­TKI a sensitive biomarker to monitor disease outcomes.14,15 treatments, as well as investigating novel therapeutic strate- Another important consideration in interpreting our study gies for patients with resistance due to p.T790 M,30,31 and the results is that mutations detected by the analysis of ctDNA paradigm of sensitive p.T790 M detection demonstrated in should be regarded as time-­dependent variables.39 In fact, LungBEAM is expected to be highly useful in these settings. our MAF progression analysis was restricted to the group of Regarding the detection of resistance to first-­line osim- patients with at least four post-­baseline plasma samples avail- ertinib treatment, it should be noted that BEAMing technol- able, so those patients with early progression were not in- ogy can also be used to detect the EGFR C797S mutation, cluded. Increasing the frequency of plasma monitoring (e.g., which is a frequent mechanism of resistance after first-­line weekly or every 2 weeks) might show even greater utility of administration of osimertinib.32 However, other resistance plasma BEAMing EGFR mutation monitoring, especially mechanisms including variations in C-­MET copy number in patients that are suspected to be at high risk for progres- variation, as well as cell cycle gene amplifications, BRAF sion. In light of this, the EGFR mutation profile in NSCLC mutations, PI3KCA mutations, and human epidermal growth patients should be determined quickly, given that the early factor receptor-­ 2 amplification/mutations,33,34 necessitate detection of circulating EGFR mutations in plasma, as well a broader coverage of these gene alterations. Hence, next-­ as increasing mutation levels, is prognostic of rapid tumor generation sequencing-­based methods may prove to be more progression, and should signal the opportunity to consider useful in identifying these cases of therapy resistance, as they transitioning to a different treatment. allow the simultaneous detection of more of these alterations Although the value of liquid biopsy to ascertain tumor mu- in multiple genes.35 However, the levels of clinical sensitivity tation status is well established,35,40,41 there are some barriers for the detection of these alterations vary considerably and to its widespread use, such as concerns about false-­negative are challenging to report with accuracy. Therefore, consider- results. In lung cancer, false-­negative results can occur more able care should be exercised in interpreting such data used frequently in plasma testing as compared to tissue testing, to support treatment recommendations.36 because the shedding of ctDNA into the blood in NSCLC Regarding the detection of EGFR sensitizing mutations, patients has been observed to be lower than in patients with the baseline and MAF levels of these mutations in plasma other solid tumors. In colorectal cancer, for instance, the con- samples obtained by BEAMing prior to the initiation of cordance of mutation results derived from plasma and tissue EGFR-­TKI therapy were significantly associated with a lower is equal to or greater than 90%.42 Nevertheless, the use of PFS, both for ex19del and L858R, even after adjusting for the BEAMing technology in LungBEAM, as well as in other other risk factors. These findings suggest that plasma EGFR studies,13,19,43 has shown high sensitivity in detecting EGFR mutation testing in combination with a tissue test at baseline mutations in plasma, which makes it a feasible alternative can provide important complementary information regard- to tissue-­based mutation analysis in order to guide therapy ing patient prognosis. Moreover, the tracking of sensitizing selection and response evaluation for NSCLC patients. mutation levels in plasma over time showed distinctly differ- ent patterns depending on whether or not the patient showed disease progression, indicating that the changing levels of 5 | CONCLUSION MAF over time are useful in the monitoring of treatment re- sponse.37,38 Patients with a persistent increase in MAF values BEAMing accurately detected the presence of EGFR muta- after the post-­baseline drop were prone to progress during tions and showed that increasing MAF values for sensitiz- the follow-­up. The patient-­specific post-­baseline MAF slope ing and p.T790 M mutations in serial plasma samples are | 20457634, 2021, 17, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4135 by Readcube (Labtiva Inc.), Wiley Online Library on [16/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 5886    GARRIDO et al. an important negative prognostic indicator. Future studies MSD, Takeda, Roche; JLGL reports consulting/advisory should be pursued to determine a relevant clinical cut-­off role honoraria from Janssen-­Cilag, MSD, BMS, Boehringer point for MAF for use as an early disease progression prog- Ingelheim, speaker honoraria from MSD, and research nostic tool, independent of imaging results, in patients with funding from Mirati Therapeutics, AstraZeneca, Bayer, metastatic NSCLC. OncoMed, Astellas Pharma, Janssen-­Cilag, Roche, Abbvie, Boehringer Ingelheim, Pfizer, PharmaMar, BMS, Novartis, ETHICAL APPROVAL STATEMENT Celgene, Ignyta; DI reports consulting honoraria from The study was approved by the institutional review board AbbVie, Amgen, AstraZeneca, BMS, Boehringer Ingelheim, at each hospital and was conducted in accordance with the Eli Lilly Oncology, F. Hoffmann-­La Roche, Merck, MSD, principles of the Declaration of Helsinki. All eligible patients Novartis, Pierre Fabre, Pfizer, Takeda, speaker honoraria signed the informed consent form prior to any study-­specific from Amgen, AstraZeneca, BMS, Boehringer Ingelheim, procedures, sampling, or analyses. Eli Lilly Oncology, F. Hoffmann-­La Roche, MSD, Novartis, Pierre Fabre, Pfizer, and research grants from AstraZeneca, ACKNOWLEDGMENTS BMS, F. Hoffmann-­La Roche, MSD, Pierre Fabre; NV re- We thank nursing and laboratory personnel, as well as data ports grants from Roche, during the conduct of the study, managers and study coordinators for their support with personal fees from Roche, other from BMS, personal fees plasma collection, plasma shipment management, and study from Pfizer, other from Boehringer Ingelheim, personal fees monitoring. We also thank Sysmex España for their techni- and other from Lilly, other from AstraZeneca, outside the cal support, Anabel Herrero, PhD, for writing assistance, and submitted work; CC has received speaker honoraria from Tracy Harrison of Springer Healthcare Communications for AstraZeneca, Roche, MSD, Pfizer, Bristol-­Myers Squibb, editorial assistance prior to submission. Takeda, for advisory role from AstraZeneca, Bristol-­Myers Squibb, MSD, Roche, Bayer, Angelini Research, and fund- CONFLICT OF INTEREST ing from AstraZeneca, Bristol-­ Myers Squibb; AI has re- JMST, AP, MLB, and JP declare no potential conflict of in- ceived honoraria for consulting or advisory role from BMS, terests. The following authors declare the following financial Boehringer Ingelheim, MSD, Pfizer, Roche, and for ex- interests/personal relationships which may be considered as pert testimony from AstraZeneca, Boehringer Ingelheim, potential competing interests: PG reports personal fees from MSD, Pfizer, Roche; OJ has received honoraria for consult- Roche, MSD, BMS, Boehringer Ingelheim, Pfizer, Abbvie, ing or advisory role from Boehringer Ingelheim, Bristol-­ Novartis, Lilly, AstraZeneca, Janssen Blueprint Medicines, Myers Squibb, Merck Sharp & Dohme, Roche/Genentech, Takeda, Gilead, Rovi, outside the submitted work; LPA has Abbvie, and research funding from Bristol-­Myers Squibb, received honorarium from Adacap, Amgen, AstraZeneca, AstraZeneca; BM reports grants and personal fees from Bayer, Blueprint Medicines, Boehringer Ingelheim, Bristol-­ Roche, personal fees and other from BMS, Takeda, other Myers Squibb, Celgene, Eli Lilly, Incyte, Ipsen, Merck, fees from MSD, Boehringer Ingelheim, outside the submit- Merck Sharp and Dohme, Novartis, Pfizer, PharmaMar, ted work; AA reports speaker honoraria from AstraZeneca Roche, Sanofi, Servier, Sysmex, Takeda, and research grant/ and for consulting/advisory role from Boehringer Ingelheim; funding from AstraZeneca, Bristol-­ Myers Squibb, Merck EF reports personal fees from Abbvie, AstraZeneca, Blue Sharp and Dohme, Pfizer; MM reports grants and personal Print Medicines, Boehringer Ingelheim, BMS, Eli Lilly, fees from BMS, personal fees and non-­financial support from Guardant Health, Janssen, Medscape, Merck KGaA, MSD, MSD, Boehringer Ingelheim, personal fees, non-­financial Novartis, Pfizer, Prime Oncology, Roche, Samsung, Takeda, support and other from AstraZeneca, Roche, personal fees Touchime, GSK, Bayer, has received grants from Grant for from Kyowa Kirin, Pierre Fabre, outside the submitted work; Oncology Innovation (GOI) and Fundación Merck Salud, TM reports speaker honoraria from Roche, BMS, Boehringer and she is an Independent Member of the Board in Grifols, Ingelheim, AstraZeneca, and has received a research grant outside the submitted work. from Kyowa Kirin; JMT reports fees for consulting/advi- sory role from MSD, BMS, Bayer, EISAI, AstraZeneca, AUTHOR CONTRIBUTION and speaker honoraria from Bayer, BMS, Takeda; JBB PG: Conceptualization, Methodology, Investigation, reports grants and personal fees from Roche-­ Genentech, Supervision, Writing -­Original Draft, Writing-­Reviewing and grants from Pfizer and Pierre Fabre, and personal fees from Editing. LPA: Conceptualization, Methodology, Investigation, MSD, BMS, AstraZeneca, and Novartis, outside the submit- Supervision, Writing -­Original Draft, Writing-­Reviewing ted work; RGC reports consulting/advisory role honoraria and Editing. MM: Investigation, Writing -­Reviewing and from AstraZeneca, Novartis, Pfizer, Boehringer Ingelheim, Editing. TM: Investigation, Writing -­Reviewing and Editing. Roche/Genentech, BMS, Takeda, MSD Oncology, Janssen JMT: Investigation, Writing -­Reviewing and Editing. JBB: Oncology, and speaker honoraria from MSD Oncology, Investigation, Writing -­Reviewing and Editing. RGC: | 20457634, 2021, 17, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4135 by Readcube (Labtiva Inc.), Wiley Online Library on [16/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License GARRIDO et al.    5887 prospective observational study. Cancer Sci. 2016;107:162-­167. Investigation, Writing -­Reviewing and Editing. JLGL: https://doi.org/10.1111/cas.12847 Investigation, Writing -­Reviewing and Editing. JMST: 8. Provencio M, Torrente M, Calvo V, et al. Dynamic circulating Investigation,Writing-­ReviewingandEditing.DI:Investigation, tumor DNA quantificaton for the individualization of non-­small-­ Writing -­Reviewing and Editing. NV: Investigation, Writing cell lung cancer patients treatment. Oncotarget. 2017;8:60291-­ -­Reviewing and Editing. CC: Investigation, Writing -­ 60298. https://doi.org/10.18632/​oncot​arget.20016 Reviewing and Editing. AI: Investigation, Writing -­Reviewing 9. Ni J, Weng L, Liu Y, Sun Z, Bai C, Wang Y. Dynamic monitoring and Editing. OJ: Investigation, Writing -­Reviewing and of EGFR mutations in circulating cell-­free DNA for EGFR-­mutant metastatic patients with lung cancer: early detection of drug resis- Editing. BM: Investigation, Writing -­Reviewing and tance and prognostic significance. Oncol Lett. 2017;13:4549-­4557. Editing. AP: Investigation, Writing -­Reviewing and Editing. https://doi.org/10.3892/ol.2017.6022 AA: Investigation, Writing -­Reviewing and Editing. 10. Lee Y, Lee GK, Lee YS, et al. Clinical outcome according to the MLB: Investigation, Writing -­Reviewing and Editing. JP: level of preexisting epidermal growth factor receptor T790M mu- Conceptualization, Methodology, Investigation, Supervision, tation in patients with lung cancer harboring sensitive epidermal Writing -­Original Draft, Writing -­Reviewing and Editing. EF: growth factor receptor mutations. Cancer. 2014;120:2090-­2098. Conceptualization, Methodology, Investigation, Supervision, https://doi.org/10.1002/cncr.28711 11. Wang S, Cang S, Liu D. Third-­generation inhibitors targeting Writing -­Original Draft, Writing -­Reviewing and Editing. EGFR T790M mutation in advanced non-­small cell lung can- cer. J Hematol Oncol. 2016;9:34. https://doi.org/10.1186/s1304​ DATA AVAILABILIT Y STATEMENT 5-­016-­0268-­z The data that support the findings of this study are available 12. Smeltzer MP, Wynes MW, Lantuejoul S, et al. The international on request from the corresponding author. The data are not association for the study of lung cancer global survey on molec- publicly available due to privacy or ethical restrictions. ular testing in lung cancer. J Thorac Oncol. 2020;15:1434-­1448. https://doi.org/10.1016/j.jtho.2020.05.002 ORCID 13. Thress KS, Brant R, Carr TH, et al. EGFR mutation detection in ctDNA from NSCLC patient plasma: a cross-­platform com- Pilar Garrido https://orcid.org/0000-0002-5899-6125 parison of leading technologies to support the clinical develop- Teresa Morán https://orcid.org/0000-0001-9172-0035 ment of AZD9291. Lung Cancer. 2015;90:509-­515. https://doi. José Luis González-­Larriba https://orcid. org/10.1016/j.lungc​an.2015.10.004 org/0000-0003-2631-0309 14. Piotrowska Z, Niederst MJ, Karlovich CA, et al. Heterogeneity underlies the emergence of EGFR T790 wild-­type clones follow- R E F E R E NC E S ing treatment of T790M-­positive cancers with a third-­generation 1. Yu HA, Arcila ME, Rekhtman N, et al. Analysis of tumor speci- EGFR inhibitor. Cancer Discov. 2015;5:713-­ 722. https://doi. mens at the time of acquired resistance to EGFR-­TKI therapy in org/10.1158/2159-­8290.CD-­15-­0399 155 patients with EGFR-­mutant lung cancers. Clin Cancer Res. 15. Karlovich C, Goldman JW, Sun J-­M, et al. Assessment of EGFR 2013;19:2240-­2247. https://doi.org/10.1158/1078-­0432.ccr-­12-­2246 mutation status in matched plasma and tumor tissue of NSCLC pa- 2. Liu Y, Sun L, Xiong ZC, et al. Meta-­analysis of the impact of de tients from a phase I study of rociletinib (CO-­1686). Clin Cancer novo and acquired EGFR T790M mutations on the prognosis of Res. 2016; 22:2386-­ 2395. https://doi.org/10.1158/1078-­ 0432. patients with non-­small cell lung cancer receiving EGFR-­TKIs. CCR-­15-­1260 Onco Targets Ther. 2017;10:2267-­2279. https://doi.org/10.2147/ 16. Oxnard GR, Thress KS, Alden RS, et al. Association between ott.s133082 plasma genotyping and outcomes of treatment with osim- 3. Huang WL, Wei F, Wong DT, Lin CC, Su WC. The emergent land- ertinib (AZD9291) in advanced non–­ small-­cell lung cancer. scape of detecting EGFR mutations using circulating tumor DNA J Clin Oncol. 2016;34:3375-­ 3382. https://doi.org/10.1200/ in lung cancer. Biomed Res Int. 2015;2015:340732. https://doi. JCO.2016.66.7162 org/10.1155/2015/340732 17. Mok TSK, Kim S-­W, Wu Y-­L, et al. Gefitinib plus chemother- 4. Sun W, Yuan X, Tian Y, et al. Non-­invasive approaches to moni- apy versus chemotherapy in epidermal growth factor receptor tor EGFR-­TKI treatment in non-­small-­cell lung cancer. J Hematol mutation-­ positive non–­small-­cell lung cancer resistant to first-­ Oncol. 2015;8:95. https://doi.org/10.1186/s1304​5-­015-­0193-­6 line gefitinib (IMPRESS): overall survival and biomarker analy- 5. Sorensen BS, Wu L, Wei W, et al. Monitoring of epidermal growth ses. J Clin Oncol. 2017;35:4027-­ 4034. https://doi.org/10.1200/ factor receptor tyrosine kinase inhibitor-­sensitizing and resistance JCO.2017.73.9250 mutations in the plasma DNA of patients with advanced non-­ 18. Elez E, Chianese C, Sanz-­García E, et al. Impact of circulating small cell lung cancer during treatment with erlotinib. Cancer. tumor DNA mutant allele fraction on prognosis in RAS-­mutant 2014;120:3896-­3901. https://doi.org/10.1002/cncr.28964 metastatic colorectal cancer. Mol. Oncol. 2019;13:1827-­1835. 6. Moran T, Felip E, Bosch-­Barrera J, et al. Monitoring EGFR-­ https://doi.org/10.1002/1878-­0261.12547 T790M mutation in serum/plasma for prediction of response to 19. Garcia J, Wozny AS, Geiguer F, et al. Profiling of circulating tumor third-­generation EGFR inhibitors in patients with lung cancer. DNA in plasma of non-­small cell lung cancer patients, monitoring Oncotarget. 2018;9:27074-­27086. https://doi.org/10.18632/​oncot​ of epidermal growth factor receptor p. T790M mutated allelic frac- arget.25478 tion using beads, emulsion, amplification, and magnetics compan- 7. Sueoka-­Aragane N, Katakami N, Satouchi M, et al. Monitoring ion assay and evaluation in futu. Cancer Med. 2019;8:3685-­3697. EGFR T790M with plasma DNA from lung cancer patients in a https://doi.org/10.1002/cam4.2244 | 20457634, 2021, 17, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/cam4.4135 by Readcube (Labtiva Inc.), Wiley Online Library on [16/02/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 5888    GARRIDO et al. 20. Sequist LV, Waltman BA, Dias-­ Santagata D, et al. Genotypic 33. Helman E, Nguyen M, Karlovich CA, et al. Cell-­free DNA next-­ and histological evolution of lung cancers acquiring resistance generation sequencing prediction of response and resistance to to EGFR inhibitors. Sci Transl Med. 2011;3:75ra26. https://doi. third-­generation EGFR inhibitor. Clin Lung Cancer. 2018;19:518-­ org/10.1126/scitr​anslm​ed.3002003 530.e7. https://doi.org/10.1016/j.cllc.2018.07.008 21. Hochmair MJ, Buder A, Schwab S, et al. Liquid-­biopsy-­based iden- 34. Ramalingam SS, Cheng Y, Zhou C, et al. Mechanisms of acquired tification of EGFR T790M mutation-­mediated resistance to afati- resistance to first-­line osimertinib: preliminary data from the nib treatment in patients with advanced EGFR mutation-­positive phase III FLAURA study. Ann Oncol. 2018;29:viii740. https://doi. NSCLC, and subsequent response to osimertinib. Target Oncol. org/10.1093/annon​c/mdy424.063 2019;14:75-­83. https://doi.org/10.1007/s1152​3-­018-­0612-­z 35. Rolfo C, Mack PC, Scagliotti GV, et al. Liquid biopsy for ad- 22. Oxnard GR, Arcila ME, Sima CS, et al. Acquired resistance to vanced non-­small cell lung cancer (NSCLC): a statement paper EGFR tyrosine kinase inhibitors in EGFR-­mutant lung cancer: from the IASLC. J Thorac Oncol. 2018;13:1248-­1268. https://doi. distinct natural history of patients with tumors harboring the org/10.1016/j.jtho.2018.05.030 T790M mutation. Clin Cancer Res. 2011;17:1616-­1622. https:// 36. Pennell NA, Arcila ME, Gandara DR, West H. Biomarker testing doi.org/10.1158/1078-­0432.ccr-­10-­2692 for patients with advanced non-­small cell lung cancer: real-­world 23. Kuiper JL, Heideman DA, Thunnissen E, et al. Incidence of T790M issues and tough choices. Am Soc Clin Oncol Educ B. 2019;531-­ mutation in (sequential) rebiopsies in EGFR-­mutated NSCLC-­ 542. https://doi.org/10.1200/EDBK_237863 patients. Lung Cancer. 2014;85:19-­24. https://doi.org/10.1016/j. 37. Thress KS, Markovets A, Barrett JC, et al. Complete clearance of lungc​an.2014.03.016. plasma EGFR mutations as a predictor of outcome on osimerti- 24. Lim C, Tsao MS, Le LW, et al. Biomarker testing and time to treat- nib in the AURA trial. J Clin Oncol. 2017;35:9018. https://doi. ment decision in patients with advanced nonsmall-­cell lung cancer. org/10.1200/JCO.2017.35.15_suppl.9018 Ann Oncol. 2015;26:1415-­1421. https://doi.org/10.1093/annon​c/ 38. Papadimitrakopoulou VA, Han JY, Ahn MJ, et al. Epidermal mdv208 growth factor receptor mutation analysis in tissue and plasma from 25. Lindeman NI, Cagle PT, Aisner DL, et al. Updated molecular test- the AURA3 trial: osimertinib versus platinum-­ pemetrexed for ing guideline for the selection of lung cancer patients for treatment T790M mutation-­positive advanced non-­small cell lung cancer. with targeted tyrosine kinase inhibitors: guideline from the col- Cancer. 2020;126:373-­380. https://doi.org/10.1002/cncr.32503 lege of American pathologists, the international association for the 39. Provencio M, Torrente M, Calvo V, et al. Prognostic value of quantita- study of lung cancer. J Mol Diagn. 2018;20:129-­159. https://doi. tive ctDNA levels in non small cell lung cancer patients. Oncotarget. org/10.1016/j.jmoldx.2017.11.004 2018;9:488-­494. https://doi.org/10.18632/​oncot​arget.22470 26. Mok T, Wu YL, Lee JS, et al. Detection and dynamic changes of 40. Hironaka-­Mitsuhashi A, Sanchez Calle A, Ochiya T, Takayama EGFR mutations from circulating tumor DNA as a predictor of sur- S, Suto A. Towards circulating-­tumor DNA-­based precision med- vival outcomes in NSCLC patients treated with first-­line intercalated icine. J Clin Med. 2019;8(9):1365 https://doi.org/10.3390/jcm80​ erlotinib and chemotherapy. Clin Cancer Res. 2015;21:3196-­3203. 91365 https://doi.org/10.1158/1078-­0432.ccr-­14-­2594 41. Bulbul A, Leal A, Husain H. Applications of cell-­free circulating 27. Johann DJ Jr, Steliga M, Shin IJ, et al. Liquid biopsy and its tumor DNA detection in EGFR mutant lung cancer. J Thorac Dis. role in an advanced clinical trial for lung cancer. Exp Biol Med. 2020;12:2877-­2882. https://doi.org/10.21037/​jtd.2020.01.66 2018;243:262-­271. https://doi.org/10.1177/15353​70217​750087 42. García-­ Foncillas J, Alba E, Aranda E, et al. Incorporating 28. El Messaoudi S, Mouliere F, Du Manoir S, et al. Circulating DNA BEAMing technology as a liquid biopsy into clinical practice for as a strong multimarker prognostic tool for metastatic colorectal the management of colorectal cancer patients: an expert taskforce cancer patient management care. Clin Cancer Res. 2016;22:3067-­ review. Ann Oncol. 2017;28:2943-­2949. https://doi.org/10.1093/ 3077. https://doi.org/10.1158/1078-­0432.CCR-­15-­0297 annon​c/mdx501 29. Grasselli J, Elez E, Caratù G, et al. Concordance of blood-­and 43. Taniguchi K, Uchida J, Nishino K, et al. Quantitative detection of tumor-­ based detection of RAS mutations to guide anti-­ EGFR EGFR mutations in circulating tumor DNA derived from lung ad- therapy in metastatic colorectal cancer. Ann Oncol. 2017;28:1294-­ enocarcinomas. Clin Cancer Res. 2011;17:7808-­7815. https://doi. 1301. https://doi.org/10.1093/annon​c/mdx112 org/10.1158/1078-­0432.ccr-­11-­1712 30. Gray JE, Okamoto I, Sriuranpong V, et al. Tissue and plasma EGFR mutation analysis in the FLAURA trial: osimertinib ver- sus comparator EGFR tyrosine kinase inhibitor as first-­line treat- SUPPORTING INFORMATION ment in patients with EGFR-­ mutated advanced non-­ small cell Additional supporting information may be found online in lung cancer. Clin Cancer Res. 2019;25:6644-­ 6652. https://doi. the Supporting Information section. org/10.1158/1078-­0432.ccr-­19-­1126 31. Rajappa S, Krishna MV, Narayanan P. Integrating osimertinib in clinical practice for non-­small cell lung cancer treatment. Adv. How to cite this article: Garrido P, Paz-­Ares L, Majem Ther. 2019;36:1279-­1290. https://doi.org/10.1007/s1232​5-­019-­ M, et al. LungBEAM: A prospective multicenter study 00917​-­6 to monitor stage IV NSCLC patients with EGFR 32. Thress KS, Paweletz CP, Felip E, et al. Acquired EGFR C797S mutations using BEAMing technology. Cancer Med. mutation mediates resistance to AZD9291 in non-­small cell lung cancer harboring EGFR T790M. Nat Med. 2015;21:560-­562. 2021;10:5878–­5888. https://doi.org/10.1002/cam4.4135 https://doi.org/10.1038/nm.3854