Radiol Oncol 2022; 56(3): 371-379. doi: 10.2478/raon-2022-0025 371 research article Real-world outcomes, treatment patterns and T790M testing rates in non-small cell lung cancer patients treated with first-line first- or second-generation epidermal growth factor receptor tyrosine kinase inhibitors from the Slovenian cohort of the REFLECT study Nina Turnsek1, Rok Devjak1, Natalija Edelbaher2, Ilonka Osrajnik2, Mojca Unk1, Dusanka Vidovic2, Tina Jeric3, Urska Janzic4 1 Department of Medical Oncology, Institute of Oncology Ljubljana, Ljubljana, Slovenia 2 Department of Pulmonary Diseases, University Medical Centre Maribor, Pohorje, Slovenia 3 AstraZeneca UK Limited, Branch office in Slovenia, Ljubljana, Slovenia 4 Department of Medical Oncology, University Clinic Golnik, Golnik, Slovenia Radiol Oncol 2022; 56(3): 371-379. Received 31 01 2022 Accepted 07 04 2022 Correspondence to: Nina Turnšek, M.D., Ph.D., Department of Medical Oncology, Institute of Oncology Ljubljana, Zaloška 2, SI-1000 Ljubljana, Slovenia. E-mail: nturnsek@onko-i.si Disclosure. NT: honoraria (self) from MSD, Boehringer Ingelheim, Roche, Pfizer, AstraZeneca; honoraria (institution) from MSD, Boehringer Ingelheim, Roche, Pfizer, AstraZeneca; RD: consulting fees from Krka, Novartis, Roche; honoraria (self) from Abbott, Amgen, AstraZeneca, Boehringer Ingelheim, Merck Sharp & Dohme, Novartis, Roche, Sanofi Aventis, Takeda; honoraria (institution) from Abbott, Amgen, AstraZeneca, Boehringer Ingelheim, Merck Sharp & Dohme, Novartis, Roche, Sanofi Aventis, Takeda; support for attendings meetings or travel, Merck Sharp & Dohme, Roche; NE: honoraria (self) from AstraZeneca, Chiesi, Pfizer, Roche, Boehringer Ingelheim, Merck Sharp & Dohme, Amgen, Eli Lilly, Novartis, Sapio, GSK, Krka; honoraria (institution) from AstraZeneca, Chiesi, Pfizer, Roche, Boehringer Ingelheim, Sapio, Merck Sharp & Dohme, Amgen, Eli Lilly, Novartis, GSK, Krka; IO: Honoraria (self) from AstraZeneca, Chiesi, Pfizer, Roche, Boehringer Ingelheim, Merck Sharp & Dohme, Amgen, Eli Lilly, Novartis, Sapio, GSK; honoraria (institution) from AstraZeneca, Chiesi, Pfizer, Roche, Boehringer Ingelheim, Sapio, Merck Sharp & Dohme, Amgen, Eli Lilly, Novartis, GSK; MU: honoraria (self) from AstraZeneca, Boehringer Ingelheim, BMS, Roche, Eli Lily, MSD, Lek, Krka, Novartis, TEVA, Pfizer; DV: honoraria (self) from AstraZeneca, Chiesi, Pfizer, Roche, Boehringer Ingelheim, Merck Sharp & Dohme, Amgen, Eli Lilly, Novartis, Sapio, GSK; honoraria (institution): AstraZeneca, Chiesi, Pfizer, Roche, Boehringer Ingelheim, Sapio, Merck Sharp & Dohme, Amgen, Eli Lilly, Novartis, GSK; TJ: employee of AstraZeneca; UJ: honoraria (self) from AstraZeneca, Boehringer Ingelheim, MSD, Roche, Pfizer; honoraria (institution) from AstraZeneca, Boehringer Ingelheim, MSD, Roche, Pfizer, Novartis, BMS This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Background. Epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) are effective treatments for EGFR mutation-positive (EGFRm) non-small cell lung cancer (NSCLC). However, routine clinical practice is different between countries/institutions. Patients and methods. The REFLECT study (NCT04031898) is a retrospective medical chart review that explored real-life treatment and outcomes of EGFRm NSCLC patients receiving first-line (1L) first-/second-generation (1G/2G) EGFR TKIs in 8 countries. This study included adult patients with documented advanced/metastatic EGFRm NSCLC with 1L 1G/2G EGFR TKIs initiated between Jan 2015 – Jun 2018. We reviewed data on clinical characteristics, treat- ments, EGFR/T790M testing patterns, and survival outcomes. Here, we report data from 120 medical charts in 3 study sites from Slovenia. Results. The Slovenian cohort (median age 70 years, 74% females) received 37% erlotinib, 32% afatinib, 31% gefitinib. At the time of data collection, 94 (78%) discontinuations of 1L TKI, and 89 (74%) progression events on 1L treatment were reported. Among patients progressing on 1L, 73 (82%) were tested for T790M mutation yielding 50 (68%) positive results, and 62 (85%) received 2L treatment. 82% of patients received osimertinib. Attrition rate between 1L and 2L was 10%. The median (95% CI) real-world progression free survival on 1L EGFR TKIs was 15.6 (12.6, 19.2) months; median overall survival (95% CI) was 28.9 (25.0, 34.3) months. Conclusions. This real-world study provides valuable information about 1G/2G EGFR TKIs treatment outcomes and attrition rates in Slovenian EGFRm NSCLC patients. The reduced attrition rate and improved survival outcomes empha- size the importance of 1L treatment decision. Key words: real-world study; non-small cell lung cancer; epidermal growth factor receptor; T790M testing, attrition Radiol Oncol 2022; 56(3): 371-379. Turnsek N et al. / Outcomes and treatments of EGFR-mutated NSCLC patients from Slovenia372 Introduction Lung cancer remains a major public health chal- lenge worldwide, due to its diagnosis in advanced stages and high rate of mortality.1,2 The discovery of sensitizing mutations to epidermal growth fac- tor receptor (EGFR) has changed the treatment paradigm for lung cancer and has allowed for im- proved outcomes in patients with tumours harbor- ing such actionable mutations.3,4 Tyrosine kinase inhibitors (TKIs) targeting EGFR have proven effi- cacy for the treatment of EGFR-mutated (EGFRm) non-small cell lung cancer (NSCLC) and are the treatment of choice when this sensitizing muta- tion is found.1 Several generations of EGFR TKIs have been developed and have become gradually available – from the first-generation erlotinib and gefitinib, to second-generation afatinib and dac- omitinib, and third-generation osimertinib.5 Initial treatment recommendations for metastat- ic EGFRm NSCLC relied on first- and second-gen- eration (1G/2G) EGFR TKIs, but despite promising initial responses to these therapies, the disease in- evitably develops resistance and the progression requires treatment change.6 In approximately half of cases, the resistance is mediated by the EGFR secondary mutation T790M1,7, which is targeted by osimertinib in exon 20.8 Based on AURA3 study results, the standard of care is now testing for the T790M mutation in all patients whose disease has progressed on 1G/2G EGFR TKIs and treatment with osimertinib when the T790M resistance muta- tion is identified.1,7,8 Based on the FLAURA study results, which showed significant survival benefit with osimertinib versus comparator EGFR TKIs, osimertinib received approval by the European Medicines Agency in 2018 and became the pre- ferred first-line (1L) treatment option in advanced or metastatic EGFRm NSCLC.1,9 The implementation of testing and treatment recommendations in clinical practice is not always a simple process. Access to new methods of mo- lecular testing and novel therapies may be affected by lengthy local approvals and reimbursement processes, particularly in Central Eastern Europe (CEE).10,11 Among countries in this region, Slovenia benefits from having a long tradition in cancer care and one of the oldest population-based cancer reg- istries in Europe.12,13 The advantage of having im- plemented a national cancer registry consists in the objective evaluation of the burden of disease and trends over time and is in direct conjunction with adequate setting and resource allocation at institu- tional level.12,14 In Slovenia, the molecular testing of EGFRm is reflex and it was partially covered by pharmaceuti- cal companies until July 2020, when it became fully reimbursed by the public health system.15 However, the reimbursement of innovative anti-cancer thera- pies is still not optimal, and it exceeds 2 years.16 For example, the newly approved osimertinib as 1L therapy was reimbursed only in October 2020. In addition to patient and tumour characteristics, the treatment decisions in real-world (RW) practice are driven by clinical and cost-effectiveness, safety, and availability of treatments.17 As shown by the recent RW experience with 1L 1G/2G EGFR TKIs, the efficacy and safety of these agents proven in registration trials usually translate in real-life prac- tice; yet, the testing rates of the resistance muta- tion T790M are not optimal.18-26 To what extent the same findings apply in the Slovenian population is unknown. For this reason, Slovenia participated in this multinational medical chart review with the overarching goals of understanding the outcomes of EGFRm NSCLC patients initiated on 1L 1G/2G EGFR TKIs, treatment and T790M testing patterns, and attrition rates in various locations from Europe and Israel.27 Here we present the results of the Slovenian patients included in this study. Patients and methods Study design and participants The retrospective medical chart review “Real- world treatment patterns, clinical outcomes, and EGFR / T790M testing practices in EGFR-mutated advanced non–small cell lung cancer patients re- ceiving First-Line EGFR TKI Therapy“ (REFLECT, ClinicalTrials.gov: NCT04031898) was conducted in 7 European countries and Israel. Overall, medi- cal chart review and data collection were carried out in 49 clinical centres from May to December 2019, and 3 comprehensive cancer care centres in Slovenia participated in this study. In Slovenia, data abstraction was conducted from October to December 2019. The study design has been reported elsewhere.27 Briefly, eligible patients for this study were ≥ 18 years of age with a confirmed diagnosis of locally advanced or metastatic EGFRm NSCLC who initi- ated 1L therapy with a 1G/2G EGFR TKI (afatin- ib, gefitinib or erlotinib) between January 1, 2015 and June 30, 2018. At the time of medical chart re- view, patients could have been alive or deceased, provided that the date of last follow-up or death was known. Patients were identified in the chrono- Radiol Oncol 2022; 56(3): 371-379. Turnsek N et al. / Outcomes and treatments of EGFR-mutated NSCLC patients from Slovenia 373 logical order of initiating 1L EGFR TKIs within the study period of interest (i.e., starting with January 1, 2015) and enrolled consecutively in the elec- tronic data collection form until the site’s quota was reached. Patients enrolled in a clinical trial for experimental treatments related to EGFRm NSCLC and patients receiving systemic treatment for their locally advanced or metastatic NSCLC prior to the 1L EGFR TKIs were excluded. In each participating country, the Institutional Review Boards (IRBs) or Ethics Committees (ECs) approved the protocol and study conduct. This medical chart review did not require informed, written consent from patients who were alive at the time of data collection unless the local IRBs/ECs required otherwise. In Slovenia, the Agency for Medicinal Products and Medical Devices (JAZMP) and the National Medical Ethics Committee (KME) approved the study, and an informed consent waiver was granted. Outcomes and definitions The primary outcome included progression events during treatment with 1L EGFR TKIs and time to progression, defined as time from initiation of 1L 1G/2G EGFR TKI therapy until the earliest sign of progression or death prior to start of a new thera- py line or start of a new therapy line. Progression was defined as radiological progression according to any imaging method, start of new therapy line, death, or other record indicative of progression, such as documented evaluation of the clinician. To differentiate this primary outcome from the progression free survival (PFS) reported in rand- omized clinical trials, we use the term “real-world PFS” (rwPFS). The secondary outcomes of this study included attrition rates and T790M testing rates among pa- tients progressing on 1L 1G/2G EGFR TKIs, types of treatments received in subsequent lines, inci- dence of central nervous system (CNS) metastases and leptomeningeal disease (LMD) and time to their development, overall survival (OS) from the start of 1L EGFR TKI therapy, and OS from first di- agnosis of CNS metastases and/or LMD to the date of death from any cause, with patients last known to be alive censored at the date of last available follow-up. Data collection Patient- and disease-specific data were obtained from the patient’s medical records and registered TABLE 1. Clinical characteristics at the time of initial NSCLC diagnosis Characteristic N = 120 n (%) Smoking history Current smoker 7 (6) Former smoker 33 (28) Never smoker 76 (63) Unknown 4 (3) ECOG performance status 0 28 (23) 1 62 (52) 2 22 (18) 3 6 (5) 4 1 (1) Unknown 1 (1) Stage at initial diagnosis Early stage (I-II) 13 (11) Limited regional (IIIA) 4 (3) Locally advanced (IIIB) 0 Metastatic (IV) 103 (86) Site of distant metastases Adrenal 12 (10) Bone 54 (45) Brain 33 (28) Liver 19 (16) Lung 60 (50) Lymph nodes 60 (50) Peritoneal 2 (2) Pleura 38 (32) Skin/soft tissue 3 (3) Other* 10 (8) * Other sites of distant metastases included: bone marrow, eye, kidney, spleen, and pericardium. ECOG=Eastern Cooperative Oncology Group by participating investigators in an electronic case report form. Each patient’s case was allocated an anonymized, encrypted identifier. Data were col- lected from the time of initial NSCLC diagnosis un- til death or the last available follow-up at the time of the patient’s inclusion in the study. Statistical analysis Sample size was based on the feasibility informa- tion received from each country, taking into ac- Radiol Oncol 2022; 56(3): 371-379. Turnsek N et al. / Outcomes and treatments of EGFR-mutated NSCLC patients from Slovenia374 count the volume of potentially eligible patients treated with 1L EGFR TKIs in the period of interest for the study. It was anticipated that each partici- pating physician would contribute with 5–30 case records to the study and each country would col- lect data from 50–180 medical records. This study had no formal statistical hypothesis; descriptive statistics were used to assess the demo- graphic and clinical characteristics, treatment pat- terns, and attrition rate. Kaplan-Meier estimators were used to describe median PFS and OS with 95% confidence interval (95% CIs). All analyses were performed in the full analysis set. The strati- fied OS analysis required > 20 number of events and a level of maturity of > 50%. The study was not powered for group comparisons. Results In total, 120 medical charts were included in this medical chart review from 3 study sites in Slovenia. The sites participating in the REFLECT study were also the only centres where lung cancer is being treated in Slovenia: 1 national cancer centre and 2 university hospitals. Demographic, clinical and EGFR mutation characteristics at baseline The median age (range) of patients was 70 (33–93) years, the majority were female (74%) and had nev- er smoked (63%). At the initial diagnosis of NSCLC, adenocarcinoma was the predominant histological subtype (99%), and the majority of patients (86%) had metastatic stage. Most patients (75%) had an Eastern Cooperative Oncology Group (ECOG) performance status of 0–1. The most frequent sites of metastases at the time of initial diagnosis were lung and lymph nodes (50% each), bone (45%), pleura (32%), and brain (28%) (Table 1). The me- dian (range) follow-up time was 24.3 (1.6–57.7) months. EGFR mutation status was determined from tis- sue biopsy (75%) or cytology specimens (25%). The specimen was extracted from the primary tumour in most cases (73%). In 2% of patients, the biopsy site was unknown. The most frequent EGFR muta- tion was exon 19 deletion (58%) followed by exon 21 L858R point mutation (28%); uncommon muta- tions (15%) included G719X, L861Q, S768I, T790M, and exon 20 insertions. First-line EGFR TKI therapy, progression and survival The 1L EGFR TKI therapies initiated during the pe- riod of interest for the study had a balanced distri- bution: 37% of patients received erlotinib, 32% ge- fitinib and 31% afatinib. At the time of data collec- tion, 94 patients (78%) discontinued 1L EGFR TKIs due to progression events or toxicities. Toxicities occurred in 9 cases (8%), with 5 of them (4%) not A B FIGURE 1. (A) Kaplan-Meier curves for median real-world progression free survival on first-line (1L) epidermal growth factor (EGFR) tyrosine kinase inhibitors (TKIs) therapy. (B) Kaplan-Meier curves for median overall survival from start of 1L EGFR TKI therapy. Censored patients are indicated with a cross. CI = confidence interval; OS = overall survival; rwPFS = real-world progression-free survival Radiol Oncol 2022; 56(3): 371-379. Turnsek N et al. / Outcomes and treatments of EGFR-mutated NSCLC patients from Slovenia 375 starting any further treatment line. A number of 26 patients (22%) continued 1L treatment. In total, 89 progression events per protocol were reported: 47 radiological progression events (39%), 22 clinical progression events (18%), 16 deaths (13%) and 4 cases (3%) with start of a new therapy line without documented progression. Median (95% CI) rwPFS was 15.6 (12.6, 19.2) months (Figure 1A). Estimated probabilities for rwPFS (95% CI) at 12, 24 and 36 months were 63% (54%, 71%), 39% (30%, 48%) and 18%% (10%, 28%), respectively. Median (95% CI) OS from start of 1L EGFR TKI was 28.9 (25.0, 34.3) months (Figure 1B). Estimated probabilities for OS (95% CI) at 12, 24 and 36 months were 83% (75%, 89%), 61% (51%, 69%) and 36% (27%, 46%), respectively. T790M mutation testing and osimertinib treatment Of the 89 patients with progression events on 1L EGFR TKI therapy, 73 (82%) were tested for the T790M mutation at any time. Of the 73 patients tested for T790M mutation, the mutation was iden- tified in 50 patients (68%) and the test was nega- tive for 23 patients (32%). Of the 73 patients with disease progression on 1L EGFR TKIs who were tested for the T790M mutation, 62 (85%) received second-line (2L) treatment. In these patients, the 2L included osimertinib (84%), chemotherapy (15%) or targeted therapy (1%). Among the rest of the 16 patients with progres- sion on 1L EGFR TKI therapy who were not tested for T790M mutation, 4 patients (25%) received 2L treatment, with either chemotherapy or osimerti- nib (50% each). Testing for the T790M mutation was performed by using liquid biopsy in most cases (77%), fol- lowed by tissue biopsy (14%) or cytology specimen (9%). Most tests (97%) were based on Cobas® EGFR mutation test (Roche). The mean time (standard deviation) between the initiation of 1L EGFR TKIs and T790M testing was 14.4 (9.0) months. Second and subsequent therapy lines Of the 89 patients with disease progression on 1L EGFR TKIs, 66 (74%) initiated 2L treatment. In the Slovenian cohort of patients, 16 (13%) patients who discontinued 1L died before receiving 2L treat- ment, while 12 (10%) patients alive of the time of 1L discontinuation did not receive any further line. The 2L treatments included osimertinib (82%), chemotherapy (17%) and other targeted therapy (1%). At the time of data collection 18 patients (28%) were still receiving 2L treatment (Figure 2). Of the 48 patients discontinuing 2L, 19 (40%) received third-line (3L) treatment, which consisted of chemotherapy (53%), targeted therapy (26%), osimertinib (16%), or immuno-oncological therapy (5%) (Figure 2). At the end of data collection, 1 pa- tient (5%) was still on 3L treatment. Attrition rates on 1L, 2L, and 3L treatment are shown in Figure 3. Of the 18 patients discontinuing 3L, 5 (28%) re- ceived fourth-line (4L) treatment, which consisted of targeted therapy (60%) or osimertinib (40%). All patients discontinued 4L, with one case of death being registered, while the remaining 4 patients received the fifth-line of treatment (5L), which con- sisted of targeted therapy (50%), chemotherapy (25%), and osimertinib (25%) (Figure 2). All pa- tients discontinued 5L treatment. Central nervous system metastases The medical charts of 46 patients (38%) recorded the presence of central nervous system (CNS) me- tastases: in 33 cases (28%) these were present at the start of 1L EGFR TKIs, and in 13 cases (11%) the CNS metastases developed after the start of 1L treatment. In all cases (100%), an imaging examina- tion (computed tomography or magnetic resonance imaging scan) was used for the diagnosis of the CNS metastases, and in 2 cases (4%) tissue biopsy was also performed. Patients with CNS metastases FIGURE 2. Treatment patterns patients in with locally advanced or metastatic epidermal growth factor receptor mutated (EGFRm) non-small cell lung cancer (NSCLC) treated with first-line (1L) first-/second-generation (1G/2G) EGFR tyrosine kinase inhibitors (TKIs). Note that multiple treatments could have been administered at each line of treatment. * Targeted therapy besides afatinib, erlotinib, gefitinib and osimertinib (1L: not specified; 3L: crizotinib); 2L = second-line; 3L =third-line; IO = immuno-oncology Radiol Oncol 2022; 56(3): 371-379. Turnsek N et al. / Outcomes and treatments of EGFR-mutated NSCLC patients from Slovenia376 had a median age (range) of 67.5 (33.0–87.0) years and most (70%) were female. Treatments applied for CNS metastases included whole brain radiation therapy (63%), targeted therapy (63%), stereotactic radiosurgery (11%) and surgical resection (9%); in 4% of cases no treatment was provided. The median (range) time from the initiation of 1L EGFR TKIs to CNS metastases diagnosed dur- ing 1L or later lines treatment was 19.8 (7.7, 34.6) months. The median (95% CI) OS in patients with CNS metastases at the start of 1L EGFR TKIs was 24.3 (18.4, 41.5) months, with 24 events reported. In the group of patients with CNS metastases devel- oped during treatment, the number of events was too small to allow reporting of OS. Leptomeningeal disease Leptomeningeal disease (LMD) was reported in 4 patients: for 1 patient before and for 3 after the start of 1L EGFR TKI therapy. In all patients the di- agnosis relied on imaging examinations only. The median (range) time from the initiation of 1L EGFR TKIs to LMD diagnosed during treatment was 19.6 (4.5, 28.7) months. The number of events was too small to allow reporting of OS. Discussion This is the first comprehensive analysis of the outcomes, treatment patterns, and testing rates in metastatic EGFRm NSCLC patients who received 1L 1G/2G EGFR TKI therapy in Slovenia over 3.5 years, from 2015 to 2018. This is a nationally rep- resentative dataset for our clinical practice because all 3 large-volume centers from Slovenia that en- sure an integrated oncology care of lung cancer pa- tients, with national coverage, participated in the REFLECT study. Considering the real-life setting, the unselected population of patients with EGFRm NSCLC and the relatively equal distribution of 1G/2G EGFR TKIs (37% erlotinib, 32% afatinib, 31% gefitinib), our findings indicate positive treatment outcomes with 1L EGFR TKIs with a median rwPFS of 15.6 months. In the overall cohort from the REFLECT study (n=896), the median rwPFS was 13.0 (95% CI 12.3, 14.1) months and more patients received afatinib (45%).27 In clinical trials of 1G/2G EGFR TKI therapy, the acquired resistance developed af- ter a median of 9.2–14.7 months of targeted treat- ment.6 Other European RW studies that partially overlap with the limits of the data collection set for the REFLECT study, but with a different distribu- tion of the 1G/2G EGFR TKI therapies have shown PFS ranging from 7.6 to 11.0 months.18-20,22,26 The en- hanced rwPFS outcomes observed in the Slovenian cohort may be the result of more standardized and homogenous cancer care across centers, including established pathways for EGFR and T790M muta- tion testing, as well as effective control policies. Furthermore, in many cases the treatment may have continued beyond radiological progression, a common approach in patients with genetic ac- tionable alterations.1 Another observational study specifically exploring the continuation of EGFR TKIs beyond radiological progression showed that patients continued treatment without clinical de- terioration for a median of 5.1 months and had a median PFS of 15.3 months.28 The median OS from the start of the 1L 1G/2G EGFR TKI therapy was 28.9 months in the Slovenian cohort and 26.2 (95% CI 23.6, 28.4) months in the overall REFLECT study cohort.27 In general, the median OS reported in RW studies with 1L 1G/2G EGFR TKIs varies greatly, due to timelines set for the analysis, factors related to the healthcare system and access to EGFR TKIs, patient characteristics and data quality. Our findings are in line with those of other reports and are relevant for the period under study, when third-generation EGFR TKI osimerti- nib was not yet approved as 1L treatment.18,20,22,24,26 Following osimertinib 1L approval and subsequent market entries, more data on the effectiveness of osimertinib in various geographies are awaited. Upon progression on 1L 1G/2G EGFR TKIs, Slovenian national guidelines for the treatment of NSCLC, in accordance with European guide- lines, recommend testing for resistance mutation FIGURE 3. Attrition rates at first-line (1L), second-line (2L) and third-line (3L) in patients with locally advanced or metastatic epidermal growth factor receptor mutated (EGFRm) non-small cell lung cancer (NSCLC). Radiol Oncol 2022; 56(3): 371-379. Turnsek N et al. / Outcomes and treatments of EGFR-mutated NSCLC patients from Slovenia 377 T790M and, in patients with positive test results initiation of osimertinib.1,29 In this cohort, 82% of patients were tested for the presence of T790M up- on progression on 1L; the resistance mutation was identified in two-thirds (68%) of these patients, thereby providing an opportunity for treatment that is effective against disease with T790M muta- tion. Expressed at the level of the overall Slovenian cohort (42%), this positive rate of T790M is in line with other RW data from European cohorts.21,23,30,31 Additionally, in most patients (n=66) receiving 2L treatment in our cohort, post-progression treat- ment consisted of osimertinib (82%), preponder- antly in patients with the T790M mutation. These results support a unified approach to T790M test- ing and subsequent treatment at the national level, consistent with guidelines recommendations.1,29 In current local practice, when a clinical progression is suspected (even before radiologic progression), an active search with minimally invasive liquid bi- opsy for the presence of resistance T790M mutation is begun. This approach allows for early initiation of 2L systemic therapy with the goal of improving patient outcomes. Over the course of the lung cancer disease, many patients develop CNS metastases, which confer a poor prognosis and present additional treatment challenges.32 CNS metastases are often identified in patients with adenocarcinoma and molecular alterations, and their incidence is sig- nificantly correlated with the presence of EGFR- activating mutations.1,33,34 In this cohort, 38% of pa- tients had CNS metastases, most of them present at the time of diagnosis of the metastatic stage of lung cancer (28%). In a local retrospective analy- sis exploring the cumulative incidence of brain metastases in 629 patients with adenocarcinomas tested for EGFRm, those with the EGFR activating mutation had a longer time to CNS progression (25.9 vs. 11.9 months, p=0.002).35 In this REFLECT study cohort, the time to CNS progression was 19.8 months, with a median OS of 24.3 months in pa- tients with CNS metastases at the start of 1L 1G/2G EGFR TKI therapy. The difference may be due to advances in radiological techniques used to iden- tify CNS metastases, as well as practice changes. The dynamic landscape of technology, improved local control and reduced morbidity are reflected in the current management of CNS metastases as stereotactic radiosurgery has become the foremost treatment modality in patients with “limited” in- tracranial disease.36 REFLECT was primarily a study of attrition rates between treatment lines. In this cohort, of the 78 patients who started 1L 1G/2G EGFR TKIs and were alive at the time of treatment discontinuation, 12 (15%) did not receive 2L treatment. The trend of not receiving further treatment was sustained in subsequent lines, although the number of pa- tients alive at the time of treatment discontinua- tion progressively decreased. The rate of patients not receiving 2L treatment after the 1L EGFR TKIs was initially reported in clinical trials and it was approximately 35%, whereas in RW studies the rate varies more widely (10–62%).25,37,38 Although the REFLECT study did not explore the reasons why patients did not receive further treatment lines, data reported in the literature suggest vari- ous causes, including lack of genetic testing, low T790M mutation rate, poor performance status and even patient’s preference not to receive the next line of treatment, which would be chemotherapy in many cases.37 In our cohort we noticed that 18% patients progressing on 1L EGFR TKI were not tested for presence of T790M mutation. The ration- ale behind the lack of T790M testing at progression was not investigated, but such finding might be explained by rapid deterioration of clinical status followed by death on 1L EGFR TKI, presence of exon 20 insertion, which is associated with limited efficacy of common EGFR TKIs and unfavorable prognosis or poor performance status at the time of disease progression rendering patient ineligible for any further systemic therapy.37,39,40 Hence, the true T790M positivity rate and proportion of pa- tients eligible for targeted 2L may be different in real-life. Beyond possible differences in healthcare setting and availability of effective treatment op- tions, exploring locally in more depth the reasons behind attrition rates is crucial to further improve patient outcomes. The real-life character of this study confers both strengths and limitations. With a minimal set of in- clusion and exclusion criteria, and a representative dataset for Slovenia, this study allowed for build- ing RW evidence on 1L 1G/2G EGFR TKI therapy at the national level based on a 3.5-year data review (2015–2018). The fact that data collection relied en- tirely on information existing in patients’ records, which sometimes have insufficient or missing data, is a key limitation in such designs. Nevertheless, Slovenia benefited from the participation of all 3 of the country’s institutions in which lung cancer is treated. As a result, data availability was very good, with minimal cases of unknown information in patients’ histories. In general, secondary data collection may be subject to selection bias, includ- ing of sites and patients. To reduce site selection bi- Radiol Oncol 2022; 56(3): 371-379. Turnsek N et al. / Outcomes and treatments of EGFR-mutated NSCLC patients from Slovenia378 as and potential patients’ spreading between sites, all 3 Slovenian comprehensive cancer centers were included in the study. To reduce patient selection bias, the ethics review package submitted has re- quested an informed consent waiver, which was granted by the National Ethics Committee. Thus, all medical records of eligible patients were consid- ered, irrespective of the vital status at the time of data collection and patients were enrolled consecu- tively in the electronic data collection form in the chronological order of starting the 1L 1G/2G EGFR TKI therapy. In contrast to clinical trials design, disease progression was not confirmed through a standardized, objective method, and the study definition reflects the RW situation (start of a new line of therapy or any other records indicative of progression, besides radiological tests). Finally, the study was not powered to compare the individual 1G/2G EGFR TKIs, and therefore outcomes could not be further characterized by molecule. Conclusions This real-world study, performed in a representa- tive dataset for Slovenian clinical practice, provides insights into the effectiveness of 1G/2G EGFR TKIs and T790M testing patterns in EGFRm NSCLC pa- tients receiving routine care. The survival outcomes and reduced attrition rate reported in this real-life setting from our country are encouraging. Newer 1L treatment options require follow-up studies to reflect the dynamic changes in clinical practice. Acknowledgments Medical writing support was provided by Ana Maria Iordan (MD, MSc) of MedInteractiv (Bucharest, Romania) and funded by AstraZeneca in accordance with Good Publication Practice (GPP3) guidelines. Statistical analyses were pro- vided by Planimeter Inc (Budapest, Hungary) and were funded by AstraZeneca. The REFLECT study (NCT04031898) was funded by AstraZeneca. Authorship: NT, RD, NE, IO, MU, DV, UJ have been involved in the acquisition and interpretation of data, critical revision of the article and approved the final version to be submitted. TJ has been in- volved in the interpretation of data, critical revi- sion of the article, supervision and approved the final version to be submitted. References 1. Planchard D, Popat S, Kerr K, Novello S, Smit EF, Faivre-Finn C, et al. Metastatic non-small cell lung cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up [published correction appears in Ann Oncol 2019; 30: 863-70]. Ann Oncol 2018; 29(Suppl 4): iv192-237. doi: 10.1093/annonc/mdy275 2. World Health Organization. International Agency for Research on Cancer (IARC). GLOBOCAN 2020. Cancer today. Lung. [internet]. [cited 2021 Feb 16]. Available at: https://gco.iarc.fr/today/data/factsheets/cancers/15- Lung-fact-sheet.pdf 3. Carcereny E, Morán T, Capdevila L, Cros S, Vilà L, de Los Llanos Gil M, et al. The epidermal growth factor receptor (EGRF) in lung cancer. Transl Respir Med 2015; 3: 1. doi: 10.1186/s40247-015-0013-z 4. Gelatti ACZ, Drilon A, Santini FC. Optimizing the sequencing of tyrosine ki- nase inhibitors (TKIs) in epidermal growth factor receptor (EGFR) mutation- positive non-small cell lung cancer (NSCLC). Lung Cancer 2019; 137: 113-22. doi: 10.1016/j.lungcan.2019.09.017 5. Vansteenkiste J, Wauters E. Tyrosine kinase inhibition of EGFR: a success- ful history of targeted therapy for NSCLC since 20 years. Ann Oncol 2018; 29(Suppl 1): i1-i2. doi: 10.1093/annonc/mdx724 6. Westover D, Zugazagoitia J, Cho BC, Lovly CM, Paz-Ares L. Mechanisms of acquired resistance to first- and second-generation EGFR tyrosine kinase inhibitors. Ann Oncol 2018; 29(Suppl 1): i10-i19. doi: 10.1093/annonc/ mdx703 7. Santoni-Rugiu E, Melchior LC, Urbanska EM, Jakobsen JN, Stricker K, Grauslund M, et al. Intrinsic resistance to EGFR-tyrosine kinase inhibitors in EGFR-mutant non-small cell lung cancer: differences and similarities with acquired resistance. Cancers 2019; 11: 923. doi: 10.3390/cancers11070923 8. Mok TS, Wu Y-L, Ahn M-J, Kim HR, Ramalingam SS, Shepherd FA, et al. Osimertinib or platinum-pemetrexed in EGFR T790M-positive lung cancer. N Engl J Med 2017; 376: 629-40. doi: 10.1056/NEJMoa1612674 9. Soria JC, Ohe Y, Vansteenkiste J, Reungwetwattana T, Chewaskulyong B, Lee KH, et al. Osimertinib in untreated EGFR-mutated advanced non-small-cell lung cancer. N Engl J Med 2018; 378: 113-25. doi: 10.1056/NEJMoa1713137 10. Cufer T, Ciuleanu TE, Berzinec P, Galffy G, Jakopovic M, Jassem J, et al. Access to novel drugs for non-small cell lung cancer in Central and Southeastern Europe: a Central European Cooperative Oncology Group analysis. Oncologist 2020; 25: e598-601. doi: 10.1634/theoncologist.2019-0523 11. Ryska A, Berzinec P, Brcic L, Cufer T, Dziadziuszko R, Gottfried M, et al. NSCLC molecular testing in Central and Eastern European countries. BMC Cancer 2018; 18: 269. doi: 10.1186/s12885-018-4023-4 12. Zadnik V, Primic Zakelj M, Lokar K, Jarm K, Ivanus U, Zagar T. Cancer burden in Slovenia with the time trends analysis. Radiol Oncol 2017; 51: 47-55. doi: 10.1515/raon-2017-0008 13. Zwitter M, Čufer T, Vrankar M, Kern I, Štupnik T, Rozman A, et al. Lung cancer in Slovenia. J Thorac Oncol 2019; 14: 1327-31. doi: 10.1016/j. jtho.2019.02.025 14. Zanetti R, Sacchetto L, Coebergh JW, Rosso S. To accelerate cancer preven- tion in Europe: challenges for cancer registries. Eur J Cancer 2018; 104: 151-9. doi: 10.1016/j.ejca.2018.09.001 15. Thunissen E, Weynand B, Udovicic-Gagula D, Brcic L, Szolkowska M, Hofman P, et al. Lung cancer biomarker testing: perspective from Europe. Transl Lung Cancer Res 2020; 9: 887-897. doi: 10.21037/tlcr.2020.04.07 16. Janzic U, Knez L, Janzic A, Cufer T. Time to access to novel anticancer drugs and the correlation with ESMO-magnitude of clinical benefit scale in Slovenia. Expert Rev Pharmacoecon Outcomes Res 2019; 19: 717-23. doi: 10.1080/14737167.2019.1702879 17. Hirsh V, Singh J. Optimal sequencing strategies in the treatment of EGFR mutation-positive non-small cell lung cancer: clinical benefits and cost- effectiveness. Am J Health Syst Pharm 2020; 77: 1466-76. doi: 10.1093/ ajhp/zxaa197 18. Leduc C, Merlio JP, Besse B, Blons H, Debieuvre D, Bringuier PP, et al. Clinical and molecular characteristics of non-small-cell lung cancer (NSCLC) harbor- ing EGFR mutation: results of the nationwide French Cooperative Thoracic Intergroup (IFCT) program. Ann Oncol 2017; 28: 2715-24. doi: 10.1093/ annonc/mdx404 Radiol Oncol 2022; 56(3): 371-379. Turnsek N et al. / Outcomes and treatments of EGFR-mutated NSCLC patients from Slovenia 379 19. Syrigos KN, Georgoulias V, Zarogoulidis K, Makrantonakis P, Charpidou A, Christodoulou C. Epidemiological characteristics, EGFR status and manage- ment patterns of advanced non-small cell lung cancer patients: the Greek REASON observational registry study. Anticancer Res 2018; 38: 3735-44. doi: 10.21873/anticanres.12654 20. Schuette W, Schirmacher P, Eberhardt WEE, Dietel M, Zirrgiebel U, Muehlenhoff L, et al. Treatment decisions, clinical outcomes, and pharma- coeconomics in the treatment of patients with EGFR mutated stage III/IV NSCLC in Germany: an observational study. BMC Cancer 2018; 18: 135. doi: 10.1186/s12885-018-4032-3 21. Reale ML, Chiari R, Tiseo M, Vitiello F, Barbieri F, Cortinovis D, et al. Be- TeaM: an Italian real-world observational study on second-line therapy for EGFR-mutated NSCLC patients. Lung Cancer 2020; 140: 71-9. doi: 10.1016/j. lungcan.2019.12.006 22. Pluzanski A, Krzakowski M, Kowalski D, Dziadziuszko R. Real-world clinical outcomes of first-generation and second-generation epidermal growth factor receptor tyrosine kinase inhibitors in a large cohort of European non- small-cell lung cancer patients. ESMO Open 2020; 5: e001011. doi: 10.1136/ esmoopen-2020-001011 23. Pereira I, Gaspar C, Pina M, Azevedo I, Rodrigues A. Real-world T790M mu- tation frequency and impact of rebiopsy in patients with EGFR-mutated ad- vanced non-small cell lung cancer. Cureus 2020; 12: e12128. doi: 10.7759/ cureus.12128 24. Gijtenbeek RGP, Damhuis RAM, Groen HJM, van der Wekken AJ, van Geffen WH. Nationwide real-world cohort study of first-line tyrosine kinase inhibitor treatment in epidermal growth factor receptor-mutated non- small-cell lung cancer. Clin Lung Cancer 2020; 21: e647-53. doi: 10.1016/j. cllc.2020.05.019 25. Chiang AC, Fernandes AW, Pavilack M, Wu JW, Laliberté F, Duh MS, et al. EGFR mutation testing and treatment decisions in patients progressing on first- or second-generation epidermal growth factor receptor tyrosine kinase inhibitors. BMC Cancer 2020; 20: 356. doi: 10.1186/s12885-020- 06826-0 26. Cuppens K, Lodewyckx L, Demedts I, Decoster L, Colinet B, Deschepper K, et al. Real-world treatment patterns, epidermal growth factor receptor (EGFR) testing and outcomes in EGFR-mutated advanced non-small cell lung cancer patients in Belgium: results from the REVEAL study. Drugs Real World Outcomes 2021; 8: 141-52. doi: 10.1007/s40801-021-00243-w 27. Addeo A, Hochmair M, Janzic U, Dudnik E, Charpidou A, Płużański A, et al. Treatment patterns, testing practices, and outcomes in the pre-FLAURA era for patients with EGFR mutation-positive advanced NSCLC: a retro- spective chart review (REFLECT). Ther Adv Med Oncol 2021; 13: 1-11. doi: 10.1177/17588359211059874 28. Goto Y, Tanai C, Yoh K, Hosomi Y, Sakai H, Kato T, et al. Continuing EGFR-TKI beyond radiological progression in patients with advanced or recurrent, EGFR mutation-positive non-small-cell lung cancer: an observational study. ESMO Open 2017; 2: e000214. doi: 10.1136/esmoopen-2017-000214 29. Unk M, Mohorčič K, Osrajnik I, Čufer T. Lung cancer – Slovenian clinical practice guidelines in systemic treatment 2017. [Slovenian]. Onkologija 2017; 21: 52-62. 30. Minari R, Mazzaschi G, Bordi P, Gnetti L, Alberti G, Altimari A, et al. Detection of EGFR-activating and T790M mutations using liquid biopsy in patients with EGFR-mutated non-small-cell lung cancer whose disease has progressed during treatment with first- and second-generation tyrosine kinase inhibi- tors: a multicenter real-life retrospective study. Clin Lung Cancer 2020; 21: e464-73. doi: 10.1016/j.cllc.2020.02.021 31. Dal Maso A, Lorenzi M, Roca E, Pilotto S, Macerelli M, Polo V, et al. Clinical features and progression pattern of acquired T790M-positive compared with T790M-negative EGFR mutant non-small-cell lung cancer: catching tumor and clinical heterogeneity over time through liquid biopsy. Clin Lung Cancer 2020; 21: 1-14.e3. doi: 10.1016/j.cllc.2019.07.009 32. Owen S, Souhami L. The management of brain metastases in non-small cell lung cancer. Front Oncol 2014; 4: 248. doi: 10.3389/fonc.2014.00248 33. Page S, Milner-Watts C, Perna M, Janzic U, Vidal N, Kaudeer N, et al. Systemic treatment of brain metastases in non-small cell lung cancer. Eur J Cancer 2020; 132: 187-98. doi: 10.1016/j.ejca.2020.03.006 34. Li L, Luo S, Lin H, Yang H, Chen H, Liao Z, et al. Correlation between EGFR mutation status and the incidence of brain metastases in patients with non-small cell lung cancer. J Thorac Dis 2017; 9: 2510-20. doi: 10.21037/ jtd.2017.07.57 35. Stanic K, Zwitter M, Hitij NT, Kern I, Sadikov A, Cufer T. Brain metastases in lung adenocarcinoma: impact of EGFR mutation status on incidence and survival. Radiol Oncol 2014; 48: 173-83. doi: 10.2478/raon-2014-0016 36. Shinde A, Akhavan D, Sedrak M, Glaser S, Amini A. Shifting paradigms: whole brain radiation therapy versus stereotactic radiosurgery for brain metastases. CNS Oncol 2019; 8: CNS27. doi: 10.2217/cns-2018-0016 37. Roeper J, Kurz S, Grohé C, Griesinger F. Optimizing therapy sequence to prevent patient attrition in EGFR mutation-positive advanced or metastatic NSCLC. Future Oncol 2021; 17: 471-86. doi: 10.2217/fon-2020-0854 38. Nadler E, Pavilack M, Espirito JL, Clark J, Fernandes A. Observational study of treatment patterns in patients with epidermal growth factor recep- tor (EGFR) mutation-positive non-small cell lung cancer after first-line EGFR-tyrosine kinase inhibitors. Adv Ther 2020; 37: 946-54. doi: 10.1007/ s12325-020-01221-4 39. Wang F, Li C, Wu Q, Lu H. EGFR exon 20 insertion mutations in non-small cell lung cancer. Transl Cancer Res 2020; 9: 2982-91. doi: 10.21037/ tcr.2020.03.10 40. Remon J, Hendriks LEL, Cardona AF, Besse B. EGFR exon 20 insertions in advanced non-small cell lung cancer: a new history begins. Cancer Treat Rev 2020; 90: 102105. doi: 10.1016/j.ctrv.2020.102105 41. Dent RAG, Cole P. In vitro maturation of monocytes in squamous carcinoma of the lung. Br J Cancer 1981; 43: 486-95. doi: 10.1038/bjc.1981.71 42. Sterne JA, Hernan MA, Reeves BC, Savovic J, Berkman ND, Viswanathan M, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016; 355: i4919. doi: 10.1136/bmj.i4919 43. Chapman S, Nakielny R. A guide to radiological procedures. London: Bailliere Tindall; 1986. 44. Evans R, Alexander P. Mechanisms of extracellular killing of nucleated mammalian cells by macrophages. In: Nelson DS, editor. Immunobiology of macrophage. New York: Academic Press; 1976. p. 45-74.