Abstract
Background/Aim: Afatinib is a standard treatment for patients with advanced non–small cell lung cancer (NSCLC) harboring epidermal growth factor receptor (EGFR) mutations. Osimertinib can overcome the treatment resistance–associated EGFR T790M mutation, and the sequence of afatinib followed by osimertinib is an effective therapeutic strategy for NSCLC patients. This study comprehensively evaluated the outcomes of sequential therapy following frontline afatinib and identified predictive factors for T790M mutation acquisition. Patients and Methods: Data from patients with advanced NSCLC treated with frontline afatinib at a Taiwanese hospital group from June 2014 to March 2018 were retrospectively reviewed. The EGFR T790M mutation was detected by tissue sequencing or liquid biopsy. The patients’ clinicopathological features were collected, and univariate and multivariate analyses were performed to identify potential predictive and prognostic factors. Results: A total of 635 patients treated with afatinib were enrolled in this study. Until August 2021, 553 patients experienced progression, and 225 patients underwent T790M mutation testing. The T790M positive rate was 54.2%. Both exon 19 deletion and progression-free survival were associated with T790M positivity. Osimertinib was found to be effective in T790M-positive but not T790M-negative NSCLC. The median overall survival (OS) was 61.8 months for patients with T790M mutation undergoing later-line osimertinib compared with 30.1 months for patients without T790M mutation undergoing chemotherapy only. Osimertinib independently prolonged OS after afatinib progression. Conclusion: This study confirmed the efficacy of sequential afatinib and osimertinib treatment. T790M mutation detection and osimertinib availability are important for prolonging survival in patients with NSCLC harboring EGFR mutations.
Prior to the development of targeted therapy, lung cancer was a difficult cancer to treat (1). The estimated number of new lung cancer cases and associated deaths ranked second and first, respectively, for both sexes when evaluated separately according to statistics from 2021 in the USA (2). Patients with early-stage lung cancer are initially treated with surgery, chemotherapy, or radiotherapy, each of which can be provided either alone or in combination (3). Even in the absence of recurrence or progression to metastatic lung cancer, systemic therapy should be the main treatment (4). Non-small cell lung cancer (NSCLC) accounts for greater than 80% of all lung cancer cases, and treatment strategies for NSCLC continue to be developed (5). Before the era of targeted therapy, platinum-based doublet chemotherapy was the standard frontline treatment strategy, which was associated with an average progression-free survival (PFS) of 4 months and overall survival (OS) of 8 months (1, 6). Nowadays, with the development of targeted therapy, the most commonly used and studied targeted agents are epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs), such as gefitinib and erlotinib [first-generation (1G)) (7–10), afatinib and dacomitinib [second-generation (2G)] (11–14), and osimertinib [third-generation (3G)] (15, 16).
The FLAURA study showed a significant improvement in OS for patients with NSCLC who harbored EGFR mutations and were treated with osimertinib as first-line therapy [median OS: 38.6 months, 95% confidence interval (CI)=34.5-41.8 months) compared with patients who received standard-of-care treatment using other EGFR-TKIs, such as gefitinib or erlotinib (median OS: 31.8 months, 95%CI=26.6-36.0 months) (15). However, the improvement in OS was marginal [hazard ratio (HR)=0.8, p=0.046) and may be limited among non-Asian populations and in patients with exon 19 deletion. In addition, the comparators were 1G EGFR-TKIs rather than 2G EGFR-TKIs, which are considered superior to 1G EGFR-TKIs (13) and are widely used in Asian populations, particularly afatinib.
Sequential treatment after afatinib is typically based on whether the acquired T790M resistance mutation is detected during progression (17). The acquired EGFR T790M mutation accounts for greater than 50% of cases that progress following treatment with 1G or 2G EGFR-TKIs (17), and osimertinib has demonstrated remarkable benefits over standard chemotherapy among patients with the acquired T790M mutation (18). However, survival outcomes among patients with NSCLC who progress on afatinib without evidence of an acquired T790M mutation or without a repeat biopsy upon progression are poorly studied. The Chang Gung Memorial Hospital (CGMH) Real-world Afatinib Treatment Experience (CREATE) study comprehensively evaluated the outcomes of sequential treatment following frontline afatinib therapy and the factors predicting the acquisition of the T790M mutation.
Patients and Methods
Data collection. Data for all patients included in this study were obtained from the Chang Gung Research Database (19), an integrated and comprehensive database consisting of multi-institutional standardized electronic medical records collected from all branches of CGMH in Taiwan, including information from the cancer registry. Patient data were obtained from the cancer registries associated with the Linkou, Kaohsiung, Keelung, and Chiayi branches of CGMH from June 2014 to March 2018.
Patient selection and clinicopathological features. Patients who were diagnosed with advanced NSCLC cancer harboring common EGFR mutations (exon 19 deletion or L858R mutation) and treated with afatinib as first-line treatment, without prior systemic treatment for advanced NSCLC, were enrolled in the study.
Clinicopathological features, including age, sex, smoking history, performance status (PS), tumor histology, tumor involvement, EGFR mutation (exon 19 deletion or L858R mutation), tumor response, and subsequent treatment, were obtained. All patients underwent tissue or liquid biopsy after progression on afatinib. A positive outcome for T790M on either tissue or liquid biopsy was considered T790M-positive. The majority of tissue biopsies were assessed using the amplification-refractory mutation system (ARMS), and a minority were assessed by next-generation sequencing. The last follow-up time point included in the study was August 2021.
Ethical considerations. This study was approved by the Institutional Review Board of CGMH (201901395B0). Patient consent to participate was not required due to the retrospective nature of this study.
Tumor response, survival, and statistical analysis. The tumor response was evaluated according to the Response Evaluation Criteria in Solid Tumors 1.1 criteria. Detailed definitions of tumor response, PFS and OS can be found in our previous study (20, 21).
Continuous variables were compared using analysis of variance (ANOVA). Categorical variables were compared using Pearson’s Chi-square test or Fisher’s exact test. PFS and OS were estimated using the Kaplan–Meier method and compared using the log-rank test. Univariate analysis was performed to evaluate potential prognostic factors, including age, sex, PS, smoking history, histology, and the location of metastases. Multivariate analysis was performed by including all variables with p<0.05 on univariate analysis to evaluate independent prognostic factors. The results are presented as the HR and 95%CI according to Cox regression analyses. IBM SPSS Statistics for Windows (Version 23.0, Armonk, NY, USA) was used to perform all statistical analyses, and p<0.05 was considered significant. Survival curves were plotted by SPSS.
Results
Sequential treatment for patients after afatinib. This study enrolled 635 patients with EGFR-mutated NSCLC treated with first-line systemic afatinib. By the end of August 2021, 553 patients experienced progressive disease, and 225 patients were evaluated for T790M mutation. Patients were divided into three groups according to sequential treatment with either 3G EGFR-TKIs or chemotherapy: osimertinib (with or without chemotherapy); 3G EGFR-TKIs other than osimertinib (with or without chemotherapy); and chemotherapy without 3G EGFR-TKIs. The classification of sequential treatment strategies and the number of patients in each group are summarized in Figure 1.
Summary of subsequential treatment strategies applied following first-line afatinib according to the presence of the T790M mutation.
Clinical factors associated with acquired T790M mutation after afatinib treatment. Among the 225 patients with known T790M mutation status, 122 patients (52.4%) were positive for T790M mutation. T790M positive rates were 47.3% (53/112) on tissue biopsy and 48.8% (66/135) on liquid biopsy. Only 22 patients were evaluated by both tissue and liquid biopsy, which typically occurred when a second biopsy and sequencing were ordered following a negative T790M mutation result in the initial examination; therefore, concordance could not be assessed in the current study. Tissue and liquid biopsies should be considered complementary, and the performance of both is likely to result in the increased detection of T790M mutations compared with the use of a single sequencing method.
The clinical factors associated with the acquisition of T790M mutation after afatinib treatment were evaluated (Table I). Patients with exon 19 deletion had higher rates of T790M mutation than those with the L858R mutation (59.4% vs. 46.0%, p=0.049, Figure 2A). In addition, longer PFS rates on first-line afatinib therapy was associated with a higher rate of T790M mutation (36.4% for <6 months, 40.4% for 6-12 months, 59.4% for 12-18 months, 60.0% for 18-24 months, and 65.4% for >24 months, p=0.032, Figure 2B). Multivariate analysis showed that PFS duration on afatinib was the only significant factor associated with T790M mutation, although exon 19 deletion showed a higher T790M mutation rate than L858R (odds ratio: 1.70, 95%CI=0.98-2.96, p=0.061; Table I).
Univariate and multivariate analyses (n=225).
T790M positivity according to EGFR mutation (A) and progression-free survival (PFS) duration on afatinib (B). (A) Patients with exon 19 deletion had higher T790M positivity than those with the L858R mutation (p=0.049). (B) Patients with longer PFS duration on afatinib had higher T790M positivity than those with shorter PFS on afatinib (p=0.032).
Treatment outcomes for later-line osimertinib therapy following frontline afatinib according to T790M mutation status. A total of 69 patients received later-line osimertinib therapy following afatinib, regardless of the presence of the T790M mutation. The numbers of patients undergoing osimertinib therapy who were identified as T790M-positive, T790M-negative, or unknown T790M status were 54, 11, and 4, respectively. The overall objective response rate (ORR) of osimertinib was 40.6%, with no significant difference observed according to T790M status (p=0.181). The overall disease control rate (DCR) was 73.9%, and patients with confirmed T790M mutation had highest DCRs among three groups (T790M-positive, T790M-negative, and with unknown T790M status) (p=0.013; Table II).
The tumor response in 69 patients who received later-line osimertinib following afatinib.
The median PFS rates on osimertinib for patients who were T790M-positive, T790M-negative, and with unknown T790M status were 11.6, 1.8, and 4.2 months, respectively (log-rank p=0.014, Figure 3A). Pairwise comparisons demonstrated that osimertinib was associated with unfavorable PFS duration in T79M-negative compared with T790M-positive patients (HR=5.23, 95%CI=1.47-18.57, p=0.011). The median OS rates measured from the initiation of first-line afatinib for patients who were T790M-positive, T790M-negative, and with unknown T790M status were 61.8, 38.1, and 31.1 months, respectively, with no significant differences between groups (log-rank p=0.143, Figure 3B). Treatment outcomes for later-line chemotherapy following frontline afatinib according to T790M mutation status
Progression-free survival (A) and overall survival (B) for patients undergoing subsequent osimertinib therapy following first-line afatinib according to the presence of the T790M mutation.
Among patients undergoing chemotherapy without 3G EGFR-TKIs following progression on afatinib, patients who are either T790M-positive or T790M-negative experience longer OS than patients with unknown T790M status (median OS: 34.6 months for T790M-positive, 30.1 months for T790M-negative, and 22.2 months for unknown T790M status, p=0.019, Figure 4). The observation of unfavorable OS among patients with unknown T790M status may reflect a rapidly progressing tumor associated with insufficient time to obtain T790M sequencing results and the unavailability of 3G EGFR-TKIs during earlier time points, leading to unfavorable OS. By contrast, no significant difference in OS was observed between patients who were T790M-positive and those who were T790M-negative who were treated with chemotherapy without 3G TKIs.
Overall survival for patients undergoing subsequent chemotherapy following first-line afatinib according to the presence of the T790M mutation.
Comparison of osimertinib and chemotherapy. T790M testing is a component of standard care in patients who progress on afatinib, and subsequent treatment should be based on T790M mutation status. The use of 3G EGFR-TKIs should be applied in patients who are T790M-positive due to little benefit to 3G EGFR-TKI treatments without T790M mutations, whereas chemotherapy is the standard treatment for patients who are T790M-negative. Therefore, survival rates between patients who are T790M-positive and treated with osimertinib were compared with those who are T790M-negative and undergoing chemotherapy without 3G EGFR-TKIs. No differences in baseline clinicopathological characteristics were observed between the two groups except for PFS duration on afatinib (p=0.026), which aligned with the finding that PFS duration was associated with T790M positivity (Table III). The median OS rates were 61.8 and 30.1 months for T790M-positive and T790M-negative groups, respectively (Figure 5).
Characteristics of patients with NSCLC treated with afatinib according to sequential treatment.
Overall survival for patients undergoing subsequent treatment following first-line afatinib. Patients with the acquired T790M mutation treated with osimertinib had longer overall survival than those without the T790M mutation undergoing chemotherapy.
Univariate analysis showed that liver metastasis, brain metastasis, number of metastatic sites, second-line treatment, and PFS duration on afatinib were associated with OS. Multivariate analysis showed that brain metastasis (HR=3.75, 95%CI=2.10-6.71, p<0.001), subsequential chemotherapy without 3G EGFR-TKIs (HR=3.29, 95%CI=1.89-5.74, p<0.001), and PFS duration were significant independent prognostic factors (Table IV). No novel prognostic factors were identified, but osimertinib was found to prolong OS in patients after progression on afatinib.
Univariate and multivariate analysis of prognostic factors for overall survival in NSCLC patients treated with afatinib followed by osimertinib or chemotherapy based on T790M mutation (n=110).
Discussion
In the current study, the overall positive rate for acquired T790M mutation after afatinib treatment in NSCLC patients harboring a common EGFR mutation was 52.4%. Patients with exon 19 deletion and those with longer PFS duration on afatinib experienced increased T790M positivity than patients with L858R mutation and those with shorter PFS duration on afatinib, respectively (Figure 2). Osimertinib resulted in a higher DCR and a longer PFS among patients with T790M compared to those without T790M (Table II, Figure 3A). Patients with T790M who receive osimertinib following afatinib therapy experienced a median OS duration longer than 5 years (61.8 months, Figure 3B). By contrast, patients undergoing chemotherapy only following afatinib, regardless of the presence of T790M mutation, experienced a median OS duration shorter than 3 years (34.6 months for T790M-positive, 30.1 months for T790M-negative, Figure 4). Therefore, the identification of patients with acquired T790M mutation who are eligible to receive afatinib–osimertinib sequential treatment could prolong OS in patients with NSCLC harboring common EGFR mutations (Figure 5).
The positive rate of acquired T790M mutation in the current study was 52.4% which aligned with previous descriptions of real-world experiences (18, 22). In addition, the PFS duration on frontline EGFR-TKI therapy and exon 19 deletion were important predictive factors for T790M acquisition in both the current study and previous reports (18, 22–24). Unfortunately, no additional predictive clinicopathological factors were identified in the current cohort. Although the T790M detection rate was higher in the AURA extension and AURA2 studies (69% for afatinib) (24) than in the current cohort, selection bias may exist in clinical trials, which has the potential to enrich the detection rate. In the report by Chiang et al. (23), repeated biopsy was found to increase the rate of T790M mutation detection, which was compatible with our findings and suggests that repeated sequencing should be considered for patients whose first biopsies show no evidence of the T790M mutations after progression.
The median OS of sequential afatinib and osimertinib therapy was 61.8 months (longer than 5 years) in the current study. The GioTag study is a global observational study for evaluating survival rates among patients with EGFR mutation-positive NSCLC treated with sequential afatinib and osimertinib therapy (25), which reported a median OS duration of 37.6 months for the entire cohort and 44.8 months for the Asian cohort. Another observational study, UpSwinG (26), reported a median OS of 36.5 months for the entire cohort and of 42.3 months among the Asian population. The median OS of sequential afatinib and osimertinib was reported as approaching 5 years in other Asian series (22, 27–29). By contrast, patients with T790M mutation undergoing chemotherapy without 3G EGFR-TKIs experienced similar OS as those without T790M mutation, which was worse than the OS of patients with T790M undergoing osimertinib treatment (Figure 3B and Figure 4). These studies, including the current one, confirmed the efficacy of sequential afatinib and osimertinib therapy in real-world practice, emphasizing the importance of detecting the acquired T790M mutation and improving the availability of osimertinib therapy.
In clinical practice, 40%-50% of patients had no T790M after afatinib treatment and were not considered suitable for osimertinib treatment. In the current study, osimertinib demonstrated little efficacy for patients without the acquired T790M mutation, although these cases were limited (n=4), likely due to an alternative resistance mechanism in this population. Chemotherapy is the standard treatment for patients without T790M mutation. Patients treated with chemotherapy without 3G EGFR-TKIs had a median OS of 30.1 months, which aligns with previous reports (22, 23, 28). Patients with the T790M mutation treated with chemotherapy without 3G EGFR-TKIs experienced similar OS durations as those without the T790M mutation in both the current cohort and in previous studies (22, 23), indicating the importance of osimertinib use in T790M-positive patients.
In conclusion, the current study comprehensively evaluated the sequential treatment in NSCLC patients harboring EGFR mutation and undergoing afatinib as frontline treatment. The real-world experience highlighted the importance of T790M mutation testing and the use of osimertinib which led to favorable survival outcomes.
Acknowledgements
This work was supported by the Chang Gung Research Database. This work was supported by grants from Linkou Chang Gung Memorial Hospital, Taoyuan, Taiwan, R.O.C. (CMRPG3J0971~3, NMRPG3K6201~3, CMRPG3K2171, CMRPG3L0911 to C-E.W.; CIRPG3H0061~2, CORPG3M0041, NMRPG3L6051 to J.W.-C.C.).
Footnotes
Authors’ Contributions
Conceptualization, C.H.C., J.W.C.C., C.Y.H., and C.E.W.; methodology, C.H.C., J.W.C.C., Y.F.F., and C.E.W.; software, J.W.C.C., Y.F.F., C.T.Y., C.H.S.K., P.C.H., and C.E.W.; investigation, C.H.C., J.W.C.C., C.Y.H., C.F.C., and C.E.W.; data curation, J.W.C.C., Y.F.F., C.T.Y., C.H.S.K., P.C.K., and C.E.W.; writing—original draft preparation, C.H.C., J.W.C.C., and C.E.W.; writing—review and editing, C.H.C., J.W.C.C., and C.E.W.; visualization, J.W.C.C., C.Y.H., C.F.C., and C.E.W.; supervision, C.E.W. All Authors have read and agreed to the published version of the manuscript.
Conflicts of Interest
The Authors declare no conflicts of interest regarding this study.
- Received January 27, 2022.
- Revision received February 14, 2022.
- Accepted February 15, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.