First-Line Tyrosine Kinase Inhibitors Combined With Local Consolidative Radiation Therapy for Elderly Patients With Oligometastatic Non-Small Cell Lung Cancer Harboring EGFR Activating Mutations

Objective The aim of this study was to investigate the efficacy and safety of combined applications of local consolidative radiation therapy (LCRT) and first-line tyrosine kinase inhibitors (TKIs) for the treatment of primary tumors and oligometastatic sites in oligometastatic NSCLC harboring Epidermal Growth Factor Receptor (EGFR) activating mutations. Patients and Methods Elderly patients with oligometastatic NSCLC (≤5 metastases) harboring EGFR activating mutations at the time of diagnosis were identified. They were treated with first-line TKIs alone or in combination with LCRT. Progression‐free survival (PFS) and overall survival (OS) were estimated through the Kaplan–Meier method. Results A total of 122 elderly patients were enrolled between February 2010 and January 2018. Among them, 41.0% (n = 50) received TKIs combined with LCRT (TKIs + LCRT group), whereas 59.0% (n = 72) received TKIs monotherapy (TKIs alone group). Patients were followed up for a median length of 34 months (ranging from 7.0 to 64 months). The median PFS in TKIs + LCRT group was 17 months (95%CI: 15.37–18.63), which was significantly longer than that of the TKIs-alone group (12 months; 95%CI: 11.05–12.95) (p <0.001). Median OS in TKIs + LCRT group was 38 months (95%CI: 35.61–40.39), while that of the TKIs-alone group was 29 months (95%CI: 26.86–31.14) (p <0.001). Multivariate analyses revealed that LCRT, one to two metastases, and good ECOG PS were independent predictors for better PFS (p <0.001, p = 0.004, and p = 0.027). Moreover, LCRT, good ECOG PS, and T1-2 stage were independent predictors for better OS (p <0.001, p = 0.007 and p = 0.007). Most of the patients suffered from grade 1 to 2 toxicities, and treatment-related deaths were not recorded. Conclusion First-line TKIs combined with LCRT may improve survival outcomes for elderly patients with oligometastatic NSCLC harboring EGFR activating mutations. This approach was not associated with much toxicity, therefore, it can be used for the treatment of elderly patients with oligometastatic disease.


INTRODUCTION
The prevalence of non-small cell lung cancer (NSCLC) among elderly people has been increasing. Approximately 50% of NSCLC patients present with distant metastases at the time of diagnosis (1). In advanced NSCLC patients with sensitizing EGFR mutations, EGFR-TKIs have been found to effectively improve clinical outcomes, relative to cytotoxic chemotherapies (2)(3)(4)(5). However, within 9-12 months of treatment, most patients develop resistance to EGFR-TKIs (6,7). In 1995, Hellman proposed the concept of "oligometastasis", which is the transition stage between local primary tumors and extensive metastasis. Several randomized clinical trials have evaluated the efficacy of different treatments for patients with oligometastatic NSCLC (8)(9)(10). However, available treatments are not effective for geriatric patients with NSCLC harboring EGFR mutations, without T790M-mediated resistance to their initial EGFR inhibitor. Unlike younger patients, elderly patients are often administered with less aggressive treatments, possibly due to resistance to EGFR-TKIs. Therefore, there is a need to develop treatment options that can suppress tumor progression among elderly patients. A combination of local consolidation therapy and EGFR-TKIs in patients with oligometastatic NSCLC harboring EGFR activating mutations significantly prolonged PFS and overall survival (OS), relative to TKIs alone (11). Currently, the efficacy and safety of combined local consolidative radiation therapy (LCRT) and first-line EGFR-TKIs in elderly patients with oligometastatic NSCLC harboring EGFR activating mutations have not been clearly defined. In this study, we tested the hypothesis that elderly patients with oligometastatic NSCLC harboring EGFR mutations may benefit from combined treatments of LCRT and first-line EGFR-TKIs.

Patients
Stage IV EGFR-mutant NSCLC patients with oligometastatic disease within 1 month of diagnosis at the Beijing Geriatric Hospital and Air Force General Hospital were retrospectively analyzed between January, 2010 and February, 2018. The inclusion criteria were: 65 years or older (elderly), pathologically confirmed NSCLC harboring EGFR-sensitizing mutations, patients with synchronous oligometastatic disease (≤5 metastases) confirmed by comprehensive imaging examinations (namely, brain magnetic resonance imaging (MRI) + whole-body positron emission tomography computed tomography (PET-CT) or brain MRI + thoracic/abdominal/pelvic CT, and bone scan when necessary), Eastern Cooperative Oncology Group Performance Status (ECOG PS) score of 2 or less, after initial treatment with first-line EGFR-TKIs, the primary tumor and all metastases were stable.
Treatment responses were assessed 6-8 weeks after treatment with first-line TKIs, in accordance with response evaluation criteria for solid tumors, Revised RECIST guidelines (version 1.1). The EGFR-TKIs used in this study were gefitinib (250 mg daily) and erlotinib (150 mg daily). Patients that responded well to first-line TKIs followed by treatment with local consolidative radiation therapy (LCRT) to primary tumor and all oligometastatic sites were designated as the TKIs + LCRT group. Patients treated with first-line TKIs alone were designated as the TKIs-alone group. Progression-free survival was defined as the time from initiation of TKIs treatment to progression or death from any cause. Overall survival was defined as the duration between the date of TKIs initiation and date of death.

Local Consolidative Radiation Therapy Procedure
The application of local consolidative radiation therapy was decided by a multidisciplinary team (namely, oncologists, radiation oncologists, radiologists, orthopedic surgeons, neurosurgeons, cardiologists, respiratory physicians, and geriatricians) at each center. Clinical assessment included the evaluation of factors such as age, cardiopulmonary functions, underlying diseases, pathological fracture risk, nutritional status, central nervous system symptoms, and risk-benefit ratio. The local consolidative radiation therapy regimen was determined by radiation oncologists based on general conditions, tumor locations, tumor sizes, tumor boundary, pulmonary functions, bone marrow, hepatic, and renal functions.

Toxicity Assessment
Acute and long-term toxicities were defined before and after 90 days using the Common Terminology Criteria for Adverse Events (CTCAE) version 5 in accordance with the National Cancer Institute (NCI). TKIs-related acute toxicities included skin rash, diarrhea, pneumonitis, neutropenia, fatigue, vomiting, and elevated ALT levels. Radiotherapy-related acute toxicities included pneumonia, fatigue, thrombocytopenia, anemia, leukopenia, dermatitis, and esophagitis.

Statistical Analysis
Statistical analyses were performed using the IBM Statistical Package for Social Sciences (SPSS) software 25 (SPSS Inc., Chicago, IL, USA), and GraphPad Prism 7 version 7.04 (GraphPad Software, Inc., La Jolla, California USA). Normally distributed quantitative data are expressed as mean and standard deviation (M ± SD). Categorical variables are presented in percentages and frequency distributions. Kaplan-Meier analysis with the log-rank test was used to calculate and analyze survival curves. A Cox proportional hazards model was used for univariate and multivariate analyses to assess possible prognostic factors and calculate survival hazard ratios (HRs) for PFS and OS at 95% confidence intervals (95%CI). The significant parameters identified in univariate analyses (p <0.05) were incorporated into multivariate Cox regression analysis to determine the independent prognostic factors. A two-sided p <0.05 was considered statistically significant.
Overall, 41% (n = 50) of the patients received local consolidative radiation therapy to primary tumor and all oligometastatic sites ( Table 2). Thirty two patients in the TKIS + LCRT group received hypofractionated radiotherapy (BED 10 ≥100 Gy) for the primary tumor (Table 3), 4 patients received gamma knife radiosurgery for isolated brain metastases, 13 patients received conventional radiotherapy plus whole-brain radiation therapy, 8 patients received gamma knife radiosurgery for contralateral lung metastases, 4 patients received hypofractionated radiotherapy for adrenal metastases, while 2 patients received hypofractionated radiotherapy for liver metastases.

Survival Outcomes
The median length of follow-up time was 34 months (range, 7.0-64 months). The median progression free survival (mPFS) time in the TKIs + LCRT group was 17 months, while that of the TKIs-alone group was 12 months (p <0.001). The median overall survival (mOS) time in the TKIs + LCRT group was 38 months, while that of the TKIs-alone group was 29 months (p <0.001) ( Figure 2). The mPFS in the entire study population was 13 months, while mOS was 34 months (95%CI: 30.3-37.7).
In the TKIs + LCRT group, clinical factors for primary tumor, BED ≥100 Gy, were associated with better OS and PFS in univariate analyses (p = 0.004 and p = 0.031, respectively) ( Figure 4). However, multivariate analysis revealed that the difference was not significant (p = 0.19 and p = 0.61, respectively).

Toxicity
For non-hematological adverse effects during EGFR-TKIs therapy, the most common events were skin rash, diarrhea, fatigue, nausea, vomiting, increased ALT levels, and pneumonitis. Most of the patients presented with grade 1-2 toxicities. Incidences of >grade 3 skin rash was 4.9% (n = 6). For radiotherapy-related acute side effects, most of the patients exhibited grade 1-2 toxicity, with very few patients exhibiting grade 3 toxicities. The incidence of grade 1-2  radiation pneumonitis in the TKIs + LCRT group was 26% (n = 13), while grade 3 pneumonia (n = 3, 6%) was only found in COPD patients. Two patients treated with head gamma knife (22 Gy/1 f) were subjected to asymptomatic radiation necrosis.

DISCUSSION
Clinical incidence for metastatic NSCLC among NSCLC patients is approximately 50%. Metastatic NSCLC is more common in  older patients than in younger patients (12). This is because, in elderly patients, the disease is at a more advanced stage at the time of diagnosis. In this study, the median age for the 1,857 metastatic NSCLC patients at the time of diagnosis was 63.5 years. Approximately 20-50% of advanced NSCLC patients were oligometastatic NSCLC (13). There is no universally accepted definition of oligometastases. However, recent studies have described oligometastases as 3-5 metastases (8-10). PET-CT has an excellent sensitivity in detecting distant and occult metastases. Detection of oligometastatic disease can be more accurate when combined with brain MRI (14,15). Approximately, 70% of patients (n = 85) received PET-CT examinations. This enabled precise determination of clinical stages of oligometastatic NSCLC. Few clinical trials have included elderly patients with advanced tumors. Therefore, it can be difficult to choose appropriate therapeutic approaches for elderly patients with advanced tumors, leading to insufficient and inappropriate treatment (16). Apart from surgery and chemotherapy, radiotherapy is one of the most important methods for cancer treatment. Very elderly patients have been shown to tolerate radiotherapy, both in definitive and palliative settings (17,18). Schmid et al. (19) reported that adoption of novel treatment approaches for the elderly population is lagging. This study was retrospective in nature, therefore, based on clinical records, we concluded that elderly patients treated with TKIs alone at the time did not receive radiation therapy because of: fear and anxiety regarding radiotherapy; lack of radiotherapy information resources; concerns regarding acute and long-term side effects; lack of symptoms of pain and a low willingness for radiation treatment. NSCLC patients with EGFR activating mutations and who are treated with TKIs exhibit longer PFS and higher response rates, compared to those with EGFR mutation-negative tumors treated with conventional chemotherapies (20,21). Elderly patients and patients with poor PS have been shown to have similar clinical benefits as younger and fitter patients, therefore, they should be offered targeted treatments in case of oncogenic driver alterations (22). However, within a year, majority of the patients eventually progress, primarily due to acquired resistance in the EGFR kinase domain. Elderly patients treated with TKIs and LCRT exhibited longer mPFS and mOS, relative to those treated with TKIs alone (17 vs.12 months, p <0.001; 38 vs. 29 months, p <0.001, respectively). Multivariate analysis showed that LCRT constitutes one of the independent favorable prognostic factors for PFS (HR = 0.32, 95%CI: 0.20-0.51, p <0.001), and OS (HR = 0.41, 95%CI: 0.27-0.63, p <0.001). These results are in accordance with recent studies, which reported that among 231 patients with EGFR mutant stage IV NSCLC, mPFS and mOS were significantly longer for the local consolidation therapy (surgery or radiotherapy) plus TKIs group than the TKIs monotherapy group (23).
Traditionally, local radiotherapy is considered to be a palliative care for elderly patients with advanced metastatic  tumors. However, treatment goals and strategies have changed from palliative care to improvement of PFS and OS. This is attributed to the introduction of the concept of oligometastatic disease. A retrospective study found that in EGFR mutant oligometastatic NSCLC patients, administration of EGFR-TKIs with local consolidation radiation therapy resulted in significantly longer mPFS, compared to TKIs monotherapy (36 vs.14 months, p = 0.0024) (24). A phase 2 randomized trial involving 29 patients reported significant benefits in PFS with addition of consolidative radiotherapy to maintenance chemotherapy for patients with oligometastatic NSCLC (9.7 vs. 3.5 months) (25). Studies have reported that patients with EGFRpositive NSCLC are more likely to develop brain metastases (26,27). In this study, 50.8% of the patients (n = 62) had brain metastases. Survival analysis did not reveal statistically significant differences in survival probabilities between patients with or without brain metastases (p = 0.342). However, in the subgroup with brain metastases, OS was significantly improved by LCRT for brain metastases (38 vs. 25 months, HR = 0.38, 95% CI: 0.21-0.69, p = 0.001). Significant comorbidities can limit life expectancy, decreasing potential survival benefits from cancer treatment (28,29). In this research, 80.3% of the elderly patients (n = 98) had underlying diseases. Only three patients died of acute myocardial infarction, with no tumor progression. Survival analysis showed that there was no evident effect of comorbidities on prognostic outcomes (HR = 1.23, 95%CI: 0.83-1.82, p = 0.300). This may be relevant in relation to how underlying diseases are controlled. In oligometastatic NSCLC, a good quality of life, a small number of metastatic sites, a small primary lung mass size and local consolidation therapy are favorable prognostic factors (30,31). In this study, good ECOG PS (HR = 0.54, 95%CI: 0.34-0.85, p = 0.007) was established to be an independent favorable prognostic factor for OS and PFS, consistent with a previous study by Sheu et al. (32). A good physical score indicated that the quality of life was satisfactory, nutritional intake was sufficient, and body immune function played a partial role. This ensured treatment continuity as the disease progressed. Moreover, we found that one to two metastases (HR = 0.57, 95%CI: 0.39-0.83, p = 0.004) were independent prognostic factors for better PFS. This implies that patients with a small number of metastatic sites have a relatively low probability for widespread metastatic disease from localized diffusion. Survival analysis revealed that LCRT to primary tumor and all oligometastatic sites is an independent favorable prognostic factor for OS and PFS, in tandem with findings from previous studies (8,33). A study reported a high local control rate in hypofractionated radiotherapy (BED 10 ≥100 Gy) lesions (34). In this study, 64% (n = 32) of the patients in the TKIS + LCRT group were subjected to hypofractionated radiotherapy (BED 10 ≥100 Gy) for primary tumor. In subgroup survival analysis (TKIs + LCRT group), patients who received BED 10 ≥100 Gy for primary tumor exhibited  (35). SRS (Gamma Knife, CyberKnife, and standard linear accelerator) for the intracranial oligometastatic disease has shown promising results. SBRT, a non-surgical alternative for oligometastatic disease, is associated with low reported toxicity and impact on the quality of life. Compared to conventional radiotherapy, SRS and SBRT have numerous advantages that increase the probability of local control while maintaining the risk of normal tissue toxicity at acceptable levels (36). Cuccia et al. (37) reported that 61 elderly patients (median age 82 years) with 90 oligometastases were treated with SBRT with a median BED10 100 Gy (range, 48-180 Gy). Local control rates at 1-and 2-years were 98.8 and 88.2%, respectively, and there were no grade 2 or higher adverse events. In this study, most patients received SBRT, stereotactic radiosurgery (SRS), or hypofractionated radiotherapy, while some patients received conventional fractionated radiotherapy. When the primary tumor was close to the esophagus or in combination with severe chronic obstructive pulmonary disease, the primary tumor was treated with conventional fractionated radiotherapy (70 Gy in 30 or 35 fractions). Most of the patients experienced grade 1-2 radiation-related esophagitis, while none of the patients developed grade 3-4 esophagitis. Vertebral metastases were treated with 30 Gy in 10 fractions or 40 Gy in 20 fractions, since the planning target volume (PTV) encompasses the entire vertebra. Meanwhile, metastasis of superficial lymph nodes was treated with conventional fractionated radiotherapy to reduce incidences of severe radiation dermatitis.
In the FLAURA study, osimertinib showed superior clinical outcomes, compared to standard EGFR-TKIs in EGFRmutated NSCLC (38). Osimertinib became a preferred first-line treatment modality in previously untreated advanced NSCLC harboring EGFR activating mutations. Zeng et al. (39) reported that 108 patients treated with osimertinib later developed oligometastatic disease. They also reported that for 14 patients who received local consolidation therapy, mPFS was significantly longer than for the osimertinib alone group (NR vs. 12.8 months, p = 0.01), and were independently associated with prolonged PFS (HR = 0.29, 95%CI: 0.12-0.68, p = 0.004). In this study, most of the patients (77.9%) received gefitinib as first-line treatment, while only 7 patients received osimertinib as subsequent treatment. This may be attributed to clinical guidelines, heavy cost of treatment, individual treatment strategies, and family treatment wishes. Approximately 50% of patients in both groups did not receive any second line treatment, including nextgeneration TKIs. This was attributed to reluctance to accept intensive chemotherapy and cost implications. For instance, before November 2018, the cost of osimertinib was mostly met by households through out-of-pocket expenditure. However, with the cost of osimertinib being covered by health insurance, more EGFR-mutated NSCLC patients received osimertinib as the first-line treatment. In this study, after initial first-line TKIs treatment, 72 patients in the TKIs alone group were not subjected to any radiotherapy in case there was no progression.
Fourteen (19.4%) patients with symptomatic brain metastases received palliative radiotherapy (WBRT 40 Gy in 20 fractions or WBRT 30 Gy in 10 fractions). Twelve (16.7%) patients received local radiotherapy for palliation of painful bone metastases (30 Gy in 10 fractions or 40 Gy in 20 fractions). This study had several limitations. i. The sample size was small, making it difficult to detect statistical differences in some subgroups. ii. It is a retrospective study, and the patients were from two centers. Patients from the Beijing Geriatric Hospital were older than those from the Air Force General Hospital (median age 73 vs. 68 years). Moreover, patients from the Beijing Geriatric Hospital had more comorbidities, compared to those from the Air Force General Hospital. Therefore, the center spends more time dealing with comorbidities before performing LCRT. Segmentation plans and doses of LCRT were also different in the two centers. Aggressive treatment approaches were performed at the Air Force General Hospital.
iii. Elderly patients were not evaluated by comprehensive geriatric assessment (CGA) and did not further clarify which type of oligometastatic NSCLC benefited more from LCRT.

CONCLUSION
LCRT for primary tumor and all oligometastatic sites in elderly patients with oligometastatic NSCLC harboring EGFR activating mutations during first-line EGFR-TKIs treatment may improve their survival outcomes with tolerable toxicity. This is a potential treatment approach for elderly patients with oligometastatic disease.

DATA AVAILABILITY STATEMENT
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Requests to access these datasets should be directed to: xiaodaigua903@163.com.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by The Ethics Committee of Beijing Geriatric Hospital.

AUTHOR CONTRIBUTIONS
Conception and design: YW, and XH. Administrative support: JZ.