Neoadjuvant EGFR-TKI Therapy for EGFR-Mutant NSCLC in Traditionally Unresectable Patients

Advanced stage non-small cell lung cancer (NSCLC) patients harboring epidermal growth factor receptor (EGFR) mutations may have benet from tyrosine kinase inhibitors (TKI). However, the role of multidisciplinary management including neoadjuvant TKI therapy and thoracic surgery is uncertain. This study assessed the possible impact of neoadjuvant TKI therapy and thoracic surgery in selected advanced stage patients. Methods Advanced stage of IIIB and IVA NSCLC patients were retrospectively reviewed from 2010 to 2013. Patients with EGFR mutations who received neoadjuvant TKI followed by surgical resection were included. All patients were followed up for 5 years or until death. Results There were total 15 advanced stage lung adenocarcinoma patients in the study. 8 patients were stage IIIB and 7 were stage IVA. All tumor sizes signicantly decreased after neoadjuvant TKI therapy (p value = 0.0002). 11 patients received adjuvant TKI therapy after surgical resection and others received adjuvant cisplatin-based chemotherapy. Progression-free survival was superior in the group of adjuvant TKI therapy than in the group of adjuvant chemotherapy (median 14 months versus 5.9 months, p value = 0.016). Overall survival (OS) was not different between two groups (p value = 0.755). In the group of adjuvant TKI therapy, median OS in patients harboring exon 19 deletion and exon 21 L858R was 60 months and 44.9 months, respectively (p value = 0.078). TKI may decrease the size of EGFR mutation lung adenocarcinoma. A multidisciplinary management including neoadjuvant TKI therapy and thoracic surgery may be discussed in selected advanced stage lung adenocarcinoma.


Introduction
Non-small cell lung cancer (NSCLC) remains the leading cause of cancer-related mortality in the developed world [1]. The 5-year relative survival rate is 18% for all stages of lung cancer and only 16% patients are localized stage with 55% of 5-year survival rate [1]. Most patients are advanced stage of III or IV with poor survival [1,2]. These patients are mainly treated with different management options, including systemic platinum-based chemotherapy, palliative radiotherapy, personalized target therapy, or immunotherapy [3][4][5]. Surgery alone is not recommended for advanced stage NSCLC patients. However, multidisciplinary management, including surgical resection in selected advanced stage NSLC patients, can improve overall survival [6-10]. Unfortunately, the use of such treatments is decreasing [7]. Page 3/15 The breakthrough identi cation of epidermal growth factor receptor (EGFR) in NSCLC in 2004 explicated the relation of overexpression of EGFR to poor prognosis has [11]. The development of ge tinib, an EGFR tyrosine kinase inhibitor (TKI), provided a notable advance in treating advanced stage EGFR-mutation positive NSCLC [12]. Following a dramatic therapeutic response, TKIs were applied in advanced stage patients to improve progression-free survival (PFS) [4,5,13]. Currently, TKIs are a rst-line treatment for advanced stage EGFR-mutation positive NSCLC patients [4,5,13]. TKIs have also been applied in localized resected NSCLC as a neoadjuvant therapy or adjuvant therapy in several clinical trials and studies [14][15][16][17].
In general, stage IIIA N2 NSCLC patients receive neoadjuvant chemotherapy followed by surgical resection in the absence of tumor progression [18]. However, neoadjuvant chemotherapy followed by surgical resection is recommended in small or single station mediastinal lymph nodes not in unresectable stage IIIA N2 NSCLC [19,20]. While thoracic surgery may possibly improve survival in advanced stage NSCLC patients and TKIs are a critical treatment for advanced stage EGFR-mutation positive NSCLC, the role of neoadjuvant TKI therapy followed by thoracic surgical resection in advanced stage EGFR-mutation positive NSCLC is still unclear. Several case reports failed to provide de nite prognosis [6, [21][22][23][24][25][26][27]. Our study investigated the possible impact of neoadjuvant TKI therapy, followed by thoracic surgery in advanced stage EGFR-mutation positive NSCLC patients.

Materials And Methods
This observational study was approved by the Institutional Review Broad of the National Taiwan University Hospital (No. 201712235RIND) and con rmed that all methods were performed in accordance with the relevant guidelines and regulations. The informed consent was waived, which was also approved by the Institutional Review Broad of the National Taiwan University Hospital (No. 201712235RIND) due to its retrospective nature. NSCLC patients with stage IIIB and IVA (eighth edition of the TNM) were retrospectively identi ed by the Department of Surgery, National Taiwan University Hospital from 2010 to 2013. Progression-free survival (PFS) and overall survival (OS) were observed for ve years.

Patient Inclusion and Exclusion Criteria
This study considered patients above the age of 20 who had initially presented with stage IIIB and IVA NSCLC patients treated with TKI therapy without tumor progression followed by surgical resection. Stage IIIB NSCLC patients without pathology proven positive N2 or N3 disease before neoadjuvant TKI therapy were excluded. Stage IVA patients without pathology proven metastases or positron emission tomography scan proven positive metastatic lesions before TKI therapy were also excluded. Patients with unknown EGFR mutant status or unknown EGFR mutant status prior to TKI treatment were excluded.
Patients without regular follow-up were also excluded.
Assessments PFS was assessed from the date of surgical resection to disease recurrence, or death from any cause, or at the end of observation. OS was assessed from the date of initial treatment to death from any cause or at the end of observation. Patients were evaluated every three to four months for two years after surgical resection and then every six months for three years. Additional evaluations were arranged if any symptoms or signs of disease recurrence were suspected. Follow-up evaluations included patient history, physical examination, chest and upper abdominal computed tomography, and brain magnetic resonance imaging. Bone scintigraphy was obtained when related symptoms occurred or bone metastasis was suspected.
Statistics SPSS version 18 (SPSS Inc., Chicago, IL, USA) was used to analyze all data. All statistical tests were two sided. Categorical variables were compared using Fisher's exact tests between two groups. Continuous variables were presented as median and range and were compared by paired t tests. The Kaplan-Meier method was used to estimate survival curves and a log-rank test was used to compare the PFS and OS between different groups. Cox regressions were also used to calculate the hazard ratios (HRs) between different groups. A p value less than 0.05 was considered to be statistically signi cant.

Results
A total of 15 advanced stage lung adenocarcinoma patients were included in the study: 2 men and 13 women with a median age of 56 years (range, 27-71). 8 patients were stage IIIB and 7 patients were stage IVA. Of the stage IVA patients, 1 patient had pleural seeding, 1 patient had adrenal gland metastasis, 4 patients had single brain metastasis, and 1 patient had single bone metastasis. Four brain metastasis patients presented with single brain tumor and initially received brain tumor excision followed by whole brain radiotherapy. The single bone metastasis patient received stereotactic ablative radiotherapy and intensity modulated radiotherapy combined with neoadjuvant TKI therapy. Two patients (13.3%) were smokers. Initially, the median tumor size in maximal diameter was 5 cm (range, 1.7-7.8). All patients received tyrosine kinase inhibitors followed by surgery. Fourteen patients (93.3%) underwent lobectomy and one patient (6.7%) received wedge resection. 2 of 15 patients received thoracotomy due to the large tumor sizes. The other patients received video-assisted thoracoscopy surgery. Of the resected tumors, the median size in maximal diameter was 2.5 cm (range, 1.0-6.6). The median follow-up time was 60.0 months (range, 28.2-60.0). Demographics and characteristics of the patients were summarized in Table  1.
As shown in Table 1, nine patients (60%) were exon 19 deletion and six patients were exon 21 L858R mutation. In terms of neoadjuvant TKI therapy, thirteen patients (86.7%) were prescribed ge tinib and two patients were prescribed erlotinib. Initially, the median tumor size was 5 cm in maximal diameter. After neoadjuvant TKI therapy, the median resected tumor size was 2.5 cm in maximal diameter. Tumor size of maximal diameter decreased after neoadjuvant TKI therapy with statistical signi cance (p value = 0.0002).
As shown in Table 2, three patients had R1 positive resection margins with initial large tumor sizes of 7.8 cm, 6.6 cm, and 5 cm in maximal diameter. No patients died within day 90 after operation. All patients received adjuvant therapies: either TKI or cisplatin-based chemotherapy. There were 11 patients continually prescribing TKI therapy after operation. Ten of 11 patients received Ge tinib. The other four patients received adjuvant chemotherapy. Fourteen patients (93.3%) developed tumor recurrences. Eight patients had intrathoracic recurrence and eight patients had extrathoracic recurrence. Brain metastasis was predominant (50%) in the extrathoracic group and recurrent lung tumor was predominant (75%) in the intrathoracic group. Ten patients were survival till the end of 5 years observation.
Of all patients, the median PFS was 9.2 months (range, 4.2 to 60.0). In the group of adjuvant TKI therapy, the median PFS was 14.0 months (range, 5.6 to 60.0). In the other group of adjuvant chemotherapy, the median PFS was 5.9 months (range, 4.2 to 9.2). Among both groups, there was signi cant difference in PFS (p value = 0.016, gure 1). Hazar ratio (HR) was 0.209 with 95% con dence interval (CI) from 0.051 to 0.849 (p value = 0.029). In the group of adjuvant TKI therapy, seven patients harbored exon 19 deletion and four patients harbored exon 21 L858R mutation. Among these two EGFR mutations, no signi cant difference was found in PFS (p value = 0.135, gure 2). HR was 0.316 with 95% CI from 0.064 to 1.55 (p value = 0.156).
In our study, we retrospectively reviewed our advanced stage EGFR-mutation positive lung adenocarcinoma patients receiving neoadjuvant TKI therapy followed by thoracic surgery. The tumor sizes decreased signi cantly after neoadjuvant TKI therapy. The difference in tumor sizes may be an effect of the high response rate of EGFR TKIs in EGFR-mutation patients [3,13]. Because tumor size is a risk factor that patients may receive thoracotomy or convert from VATS to thoracotomy [29]. The bene t of decreased tumor size may let patients receive minimally invasive thoracic surgery as feasible. Although two patients in our study still received thoracotomy due to tumors respectively measuring 5.6 cm and 6.0 cm in maximal diameter, the tumor sizes were decreased after receiving neoadjuvant TKI therapy. The most patients received VATS uneventfully. Our study showed that advanced stage EGFRmutation positive lung adenocarcinoma patients receiving neoadjuvant TKI therapy may experience decreased tumor size, allowing for the use of VATS as possible instead of thoracotomy.
Many clinical trials have shown that median PFS gradually decreases as disease progresses from early stage to advanced stage [3,[13][14][15][16][17]30]. Median PFS in the advanced stage was around ve to fourteen months associated with different types of management [3,13]. However, there is an issue that what kind of adjuvant management is feasible for the patients receiving neoadjuvant TKI therapy followed by thoracic surgery. In our study, the median PFS of all patients was 9.2 months, similar to previous clinical ndings. The mean neoadjuvant TKI interval in the group receiving adjuvant TKI therapy and adjuvant chemotherapy was 107.2 days and 73 days, respectively. The difference was not signi cant (p value = 0.7177). Our patients received adjuvant TKI therapy or adjuvant chemotherapy after operation, the median PFS in the group receiving adjuvant TKI therapy was 14.0 months. The result was signi cantly better than the group receiving adjuvant chemotherapy. Previous studies in stage II to IIIA NSCLC reported the superior results in adjuvant TKI therapy versus adjuvant chemotherapy [31]. There was also the same trend in our patients with the advanced stages of stage IIIB and IVA. The result might give us a hint that advanced stage lung adenocarcinoma patients with EGFR mutation receiving TKI as a neoadjuvant and adjuvant therapy might be an acceptable multidisciplinary management with comparable PFS. In addition, some studies have shown that patients harboring exon 19 deletion have better PFS than those with exon 21 L858R mutation, while others show different ndings [13,28,[32][33]. However, PFS showed no signi cant difference between exon 19 deletion and exon 21 L858R mutation in our study.
Generally, median OS and ve-year survival rate gradually decreased corresponding to staging [1,2,30]. In our study, the 5-year survival rate was 66.7%. OS differences between patients treated with TKI or chemotherapy were not statistically signi cant in many studies and remain uncertain [3,13,28,[32][33]. Our study also showed the same result. Although the 5-year survival rate of our patients seem to be superior to general population of advanced NSCLC patients, we could not explain where the bene t on OS coming from. Many factors involved during a whole course of treatment and in uenced a result of OS. Multidisciplinary management might be a trend to deal with advanced stage NSCLC patients. No single management would be a critical role in treating advanced stage cancer patients. Besides, OS results reported by some studies show patients harboring exon 19 deletion treated with TKI have longer survival than patients harboring exon 21 L858R mutation [34]. However, our OS results were not different between these two mutations. Not all patients received adjuvant TKI therapy and 26.7% patients received adjuvant chemotherapy. This might be a factor to in uence the result of OS in our study.

Limitation
Our study is retrospective and is limited to a small number of patients, thus being prone to have selection bias. The neoadjuvant TKI interval is not consistent nor de nitely rule-based. Our result revealed that the tumor size was decreased after neoadjuvant TKI therapy, but we could not get the conclusion how long interval should patients take neoadjuvant TKI therapy. There is also a bias in adjuvant chemotherapy. The doublet regime of adjuvant chemotherapy was cisplatin-based not totally uniform due to a retrospective study. Because our patient number was small that we could not get the conclusion in OS between two mutation statuses even the patients harboring exon 19 deletion have the trend of superiority.

Conclusion
In selected advanced stage EGFR-mutation positive adenocarcinoma patients, they may possibly have a bene t from TKI therapy in decreasing tumor size. Adjuvant TKI therapy seem to have a trend of superiority in PFS comparing to adjuvant chemotherapy. The multidisciplinary management including neoadjuvant TKI therapy, surgery, and adjuvant TKI therapy maybe an issue to discuss in advanced stage NSCLC. Future work will increase the patient sample size. A randomized control study should be performed in selected advanced stage NSCLC patients.