Radiotherapy with continued EGFR‐TKIs for oligoprogressive disease in EGFR‐mutated non‐small cell lung cancer: A real‐world study

Abstract Background Epidermal growth factor receptor (EGFR)‐mutated non‐small cell lung cancer (NSCLC) develops resistance to tyrosine kinase inhibitors (TKIs). Here, we evaluated the efficacy of radiotherapy and continuation of TKIs in patients with advanced NSCLC with oligoprogression after EGFR‐TKIs. Methods From January 2011 to January 2019, 33 patients with EGFR‐mutated NSCLC on TKIs were treated by radiotherapy and continuation of TKIs for oligoprogressive disease. The primary endpoints were median progression‐free survival 1 (mPFS1), mPFS2, and median overall survival (mOS). PFS1 was measured from the start of EGFR‐TKIs therapy to the oligoprogression of the disease. PFS2 was measured from the date of oligoprogression to the further progression of the disease, while OS was calculated from oligoprogression to death from any cause or was censored at the last follow‐up date. Result The mPFS1, mPFS2, and mOS were 11.0 (95% CI, 4.4–17.6), 6.5 (95% CI, 1.4–11.6) and 21.8 (95% CI, 14.8–28.8) months, respectively. Univariate analysis showed that EGFR mutation type (p = 0.024), radiotherapy method (p = 0.001), and performance status (p = 0.017) were significantly correlated with PFS2. Univariate analysis showed that sex (p = 0.038), smoking history (p = 0.031), EGFR mutation type (p = 0.012), and radiotherapy method (p = 0.009) were significantly correlated with OS. Multivariate analysis suggested that radiotherapy method (p = 0.001) and performance status (p = 0.048) were prognostic factors for PFS2, and radiotherapy method (p = 0.040) was a prognostic factor for OS. Conclusion Radiotherapy with continued TKIs is effective for EGFR‐mutated NSCLC with oligoprogression, and it should be conducted as soon as possible. T790M+ patients have higher sensitivity to radiotherapy, and patients with good performance status and stereotactic body radiation therapy have better PFS2 and OS.


| INTRODUCTION
Non-small cell lung cancer (NSCLC) is the most common type of lung cancer, accounting for approximately 85% of all cases. 1 One-third of patients with NSCLC showed advanced metastatic disease at the time of diagnosis. Somatic driver oncogene mutations such as epidermal growth factor receptor (EGFR) and anaplastic lymphoma kinase (ALK) have been frequently identified in NSCLC. 2 In the Asian population, 30%-40% of NSCLC patients with adenocarcinoma exhibit EGFR mutation. 3 EGFR mutation-positive advanced NSCLC and EGFR tyrosine kinase inhibitors (TKIs) treatment is associated with better survival outcomes and fewer side effects than treatment with standard first-line platinumbased chemotherapy. 4,5 However, almost all patients will progress, with a median progression-free survival (PFS) of 10-14 months and a median overall survival (OS) of 38 months.
Oligometastatic disease was first described by Hellman and Weichselbaum in 1995 as a status between metastatic systemic disease and localized disease. 6 The definition of oligometastatic disease varies, but most cases indicate fewer than five sites of metastasis. Recently, a new definition was proposed by a European expert group that included ≤5 metastases with ≤3 organs involved, and all metastatic sites could be amenable to local radical treatment. 7 Oligoprogressive sites are resistant to TKIs, however, some tumor sites might still be sensitive to TKIs, and continuation of EGFR-TKIs may provide additional benefits in patients with oligoprogression. Radiotherapy to oligoprogressive sites might restore the sensitivity of the metastatic disease and improve the survival outcome. 8,9 A retrospective matched cohort study investigated the clinical outcome of EGFR mutation-positive stage IV NSCLC receiving radiotherapy or chemotherapy for progression and the results showed that the mPFS was significantly higher in radiotherapy group than chemotherapy group (7.0 vs. 4.1, p = 0.0017), indicating that radiotherapy could extend EGFR TKI therapy for patients with oligoprogression. 10 There is limited evidence for oligometastatic radiotherapy after EGFR mutation-positive NSCLC targeted therapy, but it is unclear what kind of patients are the dominant population and the time of oligoprogression, and whether radiotherapy technology has an effect on efficacy.

| Patient selection
We conducted a retrospective analysis of NSCLC patients who developed oligoprogression during TKIs therapy between January 2011 and January 2018 at the Second Xiangya Hospital of Central South University. The study participants were registered according to the following inclusion criteria: (i) histological or cytological confirmation of advanced NSCLC; (ii) presence of tumor that harbors an EGFR mutation known to benefit from treatment with EGFR-TKIs and developed oligoprogressive disease during EGFR-TKIs therapy; (iii) treatment with continued EGFR-TKIs therapy and radiotherapy for oligoprogressive disease until further progression (radiotherapy was administered to 33 tumor sites in the brain, lungs, and bones); (iv) absence of severe comorbid conditions; (v) willingness to provide written informed consent. This study was performed in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of the National Cancer Institute and the Ethics Committees of the Second Xiangya Hospital (Ethical approval number: 202008330).

K E Y W O R D S
EGFR-mutated non-small-cell lung cancer, oligoprogression, radiotherapy, survival, tyrosine kinase inhibitors

| Follow-up and evaluation of the treatment response
The date of progression was defined based on routine surveillance imaging, which prompted earlier radiographic evaluation with routine imaging every 2 months for most patients. Routine surveillance imaging included chest and abdominal computed tomography. Patients underwent bone scan or brain magnetic resonance imaging when bone or brain metastasis was suspected, and positron emission tomography scan was performed when systemic progression was suspected. Treatment response was evaluated in accordance with the Response Evaluation Criteria in Solid Tumors (RECIST) ver. 1.1.

| Statistical analysis
The primary research points were the median progressionfree survival 1 (mPFS1), mPFS2, and median overall survival (mOS). PFS1 was measured from the start of front-line EGFR-TKIs therapy until the oligoprogression of the disease. PFS2 was measured from the date of oligoprogression until further progression of the disease. OS was calculated from oligoprogression to death from any cause or was censored at the last follow-up date. The prognostic roles of the clinical and pathologic variables were tested. The prognostic factors selected for analysis included sex (male vs. female), age (>61 years vs. ≤61 years), smoking status, performance status (PS) score (0 to 1 vs. 2), initial resectable (resectable vs. unresectable), radiotherapy method (SBRT vs. non-SBRT), EGFR mutation type (exon 19 deletion vs. exon 21 L858R mutation), number of metastatic sites (1 vs. >1), sites of radiation (brain vs. lung vs. bone), T790M status, and time of oligometastasis to radiotherapy (>1 month vs. ≤1 month). Univariate analysis was performed to analyze the possible prognostic factors. Selected factors (p < 0.10) were evaluated using multivariate analysis with the Cox regression model. In the multivariate analysis, p < 0.05 was considered statistically significant.
The T790M mutation was tested in nine patients: Five patients were positive and four patients were negative.

F I G U R E 1
The Kaplan-Meier curve for progression free survival of resistance to EGFR-TKIs is shown for the 33 patients. The median progression-free survival was 11.0 months.

| Toxicity
The toxicities observed, including skin rash, neutropenia, fatigue, anorexia, nausea, and vomiting, were controllable. No G3 to 4 toxicities were recorded, and no other severe acute or late toxicities related to this study treatment were observed.

| DISCUSSION
In the past decade, target therapy such as the EGFR tyrosine kinase inhibitors (TKIs) treatment has significantly improved the survival out advanced-stage NSCLC with driver oncogene mutations; however, almost all patients develop resistance. 4,5 The therapeutic strategy beyond progression needs to be determined according to the progression patterns. 11 For the oligoprogressive disease, discontinuation of TKIs may result in the regrowth of TKIs-sensitive clones and rapid tumor regrowth. In addition, the oligoprogressive disease might be treated with radiotherapy, whereas EGFR-TKIs can be preserved and continued. 9 In our study, patients treated with EGFR-TKIs had a median PFS1 of 11.0 months, consistent with previous studies. 9 Although it has been demonstrated that radiotherapy obtained good local control rates in NSCLC patients with oligometastatic disease at diagnosis, 9,[11][12][13] data on the efficacy of local therapy such as radiotherapy on oligoprogressive disease after target therapy is scarce. 9,14 A recent study showed that local ablative therapy (radiation or surgery) and continuation of TKIs could extend disease control by >6 months in patients with ALK+ or EGFR-MT NSCLC on erlotinib or crizotinib therapy who developed oligoprogressive disease. 9 Consistent with this study, our results also demonstrated that radiotherapy with continued TKIs is effective for EGFR-mutated NSCLC with oligoprogression.
Although radiotherapy may be effective for patients with oligoprogressive disease, the criteria for patient selection remain undetermined. In our study, the Cox regression model showed no significant correlation with PFS. The T790M mutation in exon 20 of the EGFR gene and MET amplification has been reported to be associated with EGFR-TKIs resistance in NSCLC patients. 15 Oligo-clones, that are resistant to EGFR-TKIs, may occur in oligoprogressive sites. 16 Our study showed that the T790M mutation was common in patients with oligoprogression and  18 indicating that osimertinib was a good choice for EGFR T790M-positive NSCLC patients who failed TKIs treatment.
Our study had several limitations. Osoegawa A et al. demonstrated that frequent EGFR structural changes could occur after a long-term history of EGFR-TKIs therapy and a repeated tissue biopsy was useful in identifying the potential resistance mechanisms. 16 Thus, it is recommended to make a rebiopsy to determine the genetic mutation status after TKIs resistance to guide further treatment. If tumor tissue was not available for re-biopsy, a liquid biopsy could be performed. The sample size was relatively small and the present study was a retrospective study. In addition, our study lacked a comparator group and a consistent definition of oligoprogression. Despite these limitations, our study indicates that radiotherapy followed by continued treatment with EGFR-TKIs is a safe and effective option for NSCLC patients with oligoprogression.

| CONCLUSION
Radiotherapy with continued EGFR-TKIs is an effective treatment for advanced NSCLC patients with oligoprogressive disease after first-line EGFR-TKIs. Multivariate analysis indicated that SBRT and PS of 0 to 1 were important prognostic factors for PFS2, and SBRT was an independent prognostic factor for OS. Moreover, patients with T790M mutation or radiotherapy within 1 month after oligometastasis had a better survival outcome. However, this difference was not statistically significant. Larger sample sizes and prospective and multicenter studies are needed to validate these clinical results in the future.

AUTHORS' CONTRIBUTIONS
Fang Wu designed the study. Chunhong Hu guided the study. Sixuan Wu selected patients and wrote the manuscript. Yuanqiang Wu and Renfang Deng assisted in manuscript revision. Yue Pan and Long Shu analyzed and interpreted the data. All authors approved the manuscript.