Gefitinib Plus Chemotherapy vs Gefitinib Alone in Untreated EGFR-Mutant Non–Small Cell Lung Cancer in Patients With Brain Metastases

Key Points Question What are the efficacy and safety of gefitinib plus chemotherapy in patients with untreated epidermal growth factor receptor (EGFR) mutated non–small cell lung cancer (NSCLC) brain metastases? Findings In this randomized clinical trial including 161 patients with untreated EGFR-mutant NSCLC brain metastases, gefitinib plus chemotherapy significantly improved intracranial progression-free survival, overall progression-free survival, and overall survival than gefitinib alone, with manageable adverse events. Gefitinib plus chemotherapy also had better intracranial, extracranial, and overall response rates than gefitinib alone. Meaning The findings of this randomized clinical trial suggest that gefitinib plus chemotherapy may be a viable first-line treatment for patients with brain metastases associated with EGFR-mutant NSCLC.


Sponsor
Sun yat-sen university cancer center 2) Patients who confirmed as having brain metastases by enhanced brain magnetic resonance imaging (MRI); asymptomatic brain metastases or stable brain metastases after mannitol or corticosteroid treatment; 3) Patients who had never received therapy (including chemotherapy, WBRT, EGFR-TKI and EGFR monoclonal antibody) after diagnosed brain metastases; 4) Patients had at least three metastatic lesions in brain, or patients with 1-2 intracranial lesions who were not suitable for brain radiotherapy, or patients with 1-2 intracranial lesions who refused brain radiotherapy, at least one intracranial lesion with the longest diameter of >5 mm; 2) The patient has symptoms of intracranial hypertension that cannot be relieved by dehydration treatment, such as nausea or vomiting (except due to factors such as chemotherapy or cerebral infarction); headache; limb movement disorders; epilepsy; cognitive or emotional disorders. 10) Previous radiotherapy to less than 25% of the bone marrow was permitted; however, previous radiotherapy to the whole pelvis was not permitted. Prior radiation therapy must have been completed at least 2 weeks prior to the first study drug treatment. Patients must recover from acute toxic effects before first study drug treatment.
11) Malignancy other than NSCLC within 5 years prior to randomization, except for adequately treated cervical carcinoma in situ, basal cell or squamous cell skin cancer, localized prostate cancer after radical resection, ductal in situ after radical resection cancer.
Patients continued treatment until intracranial progressive diseases, unacceptable adverse, or any cause of death.

Safety assessment
All adverse events during study period are monitored, including clinical symptoms, vital signs and laboratory examination abnormalities, with clinical manifestation, severity, onset time, duration, management and outcome recorded, and its correlation with study drugs is identified.
Assessments of safety are made according to NCI-CTCAE 4.0.

Statistical methods
For primary endpoint (iPFS), Kaplan-Meier method was used to estimate the median value and its 95% CI, and the survival graph was depicted. Log-Rank was used to test the survival difference between the two groups. Cox regression models will be applied to estimate differences between treatment groups adjusted for selected prognostic factors (such as sex, age, smoking, mutation type, presence or absence of central nervous system symptoms).
For secondary efficacy endpoints, PFS,and OS are generated by Kaplan-Meier method with estimation for median and 95% CI. Survival differences are compared between groups with stratified log-rank test.
The comparison of intracranial ORR,total ORR between two group are evaluated using Fisher's exact test.
The incidence of adverse events between two groups are compared using Fisher's exact test.

Background
Lung cancer is one of the most common malignant tumors, and brain metastasis is an important cause of treatment failure and death in patients with lung cancer [1]. The prognosis of brain metastases from lung cancer is poor, and the treatment effect is unsatisfactory, whether it is single or multiple [2]. The incidence of brain metastases in non-small cell lung cancer (NSCLC) is about 30-50%, of which about 25-30% of newly diagnosed NSCLC patients have brain metastases [3]. Studies have shown that patients with lung adenocarcinoma and patients with EGFR gene mutations are prone to brain metastases [4]. In recent years, with the increase in the incidence of lung cancer, the improvement of diagnostic technology and the treatment efficacy of tumors, the overall survival of patients has been prolonged, and the incidence of NSCLC brain metastases has been increasing.
Despite advances in the treatment of NSCLC, brain metastases are still an important factor leading to short survival and poor quality of life. There is no effective treatment for NSCLC brain metastases, and the prognosis is poor [5]. The median survival time of untreated patients is less than 3 months, and the 1-year natural survival rate is less than 10%. Whole brain radiotherapy (WBRT) dominates the treatment of brain metastases, and WBRT can prolong the median survival time of patients to 3-6 months [6,7]. Whole brain radiotherapy 30Gy/10 times is currently the generally accepted radiotherapy regimen. Single brain metastases can be treated with surgery or stereotactic radiation therapy (SRS). However, most patients with brain metastases from lung cancer cannot receive surgery and SRS because of multiple intracranial lesions, and WBRT has dose limitations, so systemic therapy can be used as a means to control intracranial and extracranial lesions. In the past 10 years, with the emergence of new drugs for the treatment of lung cancer, including pemetrexed, paclitaxel, docetaxel, vinorelbine, gemcitabine, etc., these drugs combined with platinum-based regimens have improved the efficacy of non-small cell lung cancer to more than 40% [8].
The survival time was prolonged, at the same time, the treatment efficacy of brain metastases was improved, and the effective rate of cisplatin-based combination regimen in the treatment of NSCLC brain metastases was 16%-50%. Brain metastases partially destroy the blood-brain barrier. The median survival (MST) of patients with brain metastases who receive chemotherapy alone is about 4-10 months, and chemotherapy combined with MST is about 6-10 months [8][9][10]. Therefore, there is still a lack of effective treatments for patients with brain metastases from NSCLC. The emergence of targeted drugs has become a new option in addition to traditional treatment methods such as radiotherapy and chemotherapy.
In recent years, the emergence of epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (EGFR-TKI) has brought the treatment of non-small cell lung cancer to a new level. Targeted therapy can make advanced non-small cell lung cancer from chemotherapy alone the median survival of 8-10 months improved to nearly 2 years in sensitive patients, represented by gefitinib and erlotinib [11,12]. EGFR-TKI efficacy is correlated with EGFR mutation status. The EGFR mutation rate in NSCLC is 10% in the Caucasian population and 30-40% in the Asian population. EGFR mutations frequently occur in women, adenocarcinoma, non-smoking, and Asian populations.
After first-line treatment with gefitinib in patients with EGFR mutated NSCLC, the objective response rate is 55%-82%, and the median progression-free survival is 8. clinical studies and found that treatment with pemetrexed in first-or second-line therapy reduced the risk of brain metastases being the first site of progression. Brain metastases occurred in 3.2% of patients who received pemetrexed and 6.6% of patients who did not receive pemetrexed. This study suggests that pemetrexed has a good efficacy on the prevention of brain metastases. Bearz et al. reported that 39 patients were enrolled in the clinical trial and received pemetrexed as second-line or above treatment, and the overall lesion control rate and brain lesion control rate were 69%.
Surprisingly, the control rate of brain lesions was 69%. The rate reached 82%, only 7 patients had brain lesions progression, and the OS was 10 months [17]. Barlesi et al. reported that 43 non-small cell lung cancer patients with brain metastases were treated with pemetrexed/cisplatin as first-line treatment. The response rates of intracranial and extracranial lesions were 41.9% and 34.9%, respectively, and the median survival time was 7.4 months. The PFS was 4.0 months [18]. In general, the pemetrexed/platinum regimen for the patients with non-squamous non-small cell lung cancer with brain metastases is well tolerated, and has a higher effective rate compared with other thirdgeneration chemotherapeutic drugs.
The newly reported FAST-ACT II clinical study [19] showed that compared with gemcitabine/cisplatin chemotherapy alone, gemcitabine/cisplatin combined with erlotinib can improve progression-free survival in the overall population, from 7.4 months to 10 months. Among them, the PFS of the EGFR mutated subgroup increased from 6.9 months to 16.8 months, and the OS increased from 20.6 months to 31.4 months.
The proportion of the second-line erlotinib treatment in the chemotherapy group alone was 77% in the overall population. The crossover rate for the mutant subgroup was 83%. Preclinical studies have demonstrated that the combination therapy of erlotinib and pemetrexed has a strong synergistic effect in 6 non-small cell lung cancer cell lines with different molecular properties, supporting the above results [20,21]. Therefore, we hypothesized that in patients with EGFR-mutated NSCLC and brain metastases, gefitinib combined with pemetrexed/platinum chemotherapy could improve PFS and OS compared with gefitinib alone. We performed this clinical trial to investigate the efficacy and safety of gefitinib plus pemetrexed/platinum chemotherapy in untreated EGFR mutated NSCLC with brain metastases.

Study objectives
The incidence of brain metastases in non-small cell lung cancer is 30% to 50%. Whole brain radiotherapy combined with chemotherapy is the conventional treatment for nonsmall cell lung cancer. As a small molecule targeted drug, EGFR-TKI has good cell penetration and can partially penetrate the blood-brain barrier. The 2013 NCCN guidelines pointed out that erlotinib has a certain efficacy on the treatment of brain metastases, and EGFR-TKI has become new options beyond traditional chemotherapy in patients with brain metastases. Therefore, in recent years, researchers are constantly looking for the optimal treatment regimes for non-small cell lung cancer with brain

Primary endpoint
Comparison of intracranial progression-free survival (iPFS) in gefitinib plus chemotherapy group and gefitinib alone group.

Secondary endpoints
(1). Comparison of progression-free survival (PFS) in gefitinib plus chemotherapy group and gefitinib alone group.
(2). Comparison of overall survival (OS) in gefitinib plus chemotherapy group and gefitinib alone group.
(3). Comparison of intracranial objective response rate (iORR) in gefitinib plus chemotherapy group and gefitinib alone group.
(4). Comparison of objective response rate (ORR) in gefitinib plus chemotherapy group and gefitinib alone group.
(5). Comparison of safety in gefitinib plus chemotherapy group and gefitinib alone group.

Study design
This study is a randomized, open-label, controlled, multicenter Phase III study. Arm A: gefitinib alone (250mg once daily); Arm B: gefitinib (250mg once daily) plus pemetrexed-platinum chemotherapy (pemetrexed 500mg/m2 combined with cisplatin 75mg/m2 or nedaplatin 80mg/m2) in a 4-week cycle, for four to six cycles, followed by gefitinib plus pemetrexed maintenance. Patients continued treatment until intracranial progressive diseases, unacceptable adverse, or any cause of death.

Randomization
The eligible patients were randomly assigned (1:1) by a computer-generated randomization sequence to receive gefitinib alone or gefitinib plus pemetrexedplatinum chemotherapy.

Study population
This study is conducted in treatment-naïve patients with confirmed NSCLC harboring EGFR mutation and brain metastases. Asymptomatic brain metastases or stable brain metastases after mannitol or corticosteroid treatment.

Inclusion criteria
1) Paitents who was histologically or cytologically confirmed non-small cell lung cancer with EGFR sensitive mutation (exon 19 deletion or 21 L858R mutation); 2) Patients who confirmed as having brain metastases by enhanced brain magnetic resonance imaging (MRI); asymptomatic brain metastases or stable brain metastases after mannitol or corticosteroid treatment; 3) Patients who had never received therapy (including chemotherapy, WBRT, EGFR-TKI and EGFR monoclonal antibody) after diagnosed brain metastases; 4) Patients had at least three metastatic lesions in brain, or patients with 1-2 intracranial lesions who were not suitable for brain radiotherapy, or patients with 1-2 intracranial lesions who refused brain radiotherapy, at least one intracranial lesion with the longest diameter of >5 mm; Creatinine clearance ³ 60ml/min (calculated according to Cockcroft-gault formula). 9) Women of childbearing age must take a pregnancy test (serum or urine) within 7 days before enrollment, and the result is negative, and are willing to use appropriate methods of contraception during the test and within 8 weeks after the last administration of the test drug. For males, consent to use an appropriate method of contraception or be surgically sterilized during the trial and within 8 weeks after the last dose of the trial drug; 10) Ability to comply with research and follow-up procedures; 11) The patient understands and voluntarily signs the written informed consent.

Exclusion criteria
Anyone who met any of the following exclusion criteria was not eligible for this trial: 1) Patients diagnosed with stage IV or recurrent non-squamous non-small cell lung cancer with brain metastases receiving any systemic chemotherapy, immunotherapy or biological therapy (eg targeted therapy such as erlotinib or gefitinib). And folic acid, vitamin B12, and dexamethasone could not be taken as required by the trial protocol.
2) The patient has symptoms of intracranial hypertension that cannot be relieved by dehydration treatment, such as nausea or vomiting (except due to factors such as chemotherapy or cerebral infarction); headache; limb movement disorders; epilepsy; cognitive or emotional disorders.

4) Previous interstitial lung disease, drug-induced interstitial disease, radiation
pneumonitis requiring hormone therapy, or any active interstitial lung disease with clinical evidence, and idiopathic pulmonary fibrosis was found on CT scan at baseline; 10) Previous radiotherapy to less than 25% of the bone marrow was permitted; however, previous radiotherapy to the whole pelvis was not permitted. Prior radiation therapy must have been completed at least 2 weeks prior to the first study drug treatment.
Patients must recover from acute toxic effects before first study drug treatment.
11) Malignancy other than NSCLC within 5 years prior to randomization, except for adequately treated cervical carcinoma in situ, basal cell or squamous cell skin cancer, localized prostate cancer after radical resection, ductal in situ after radical resection cancer.

Treatment period examination
1) Evaluation of adverse events (AEs): Evaluations were performed every 8 weeks until tumor progression or discontinuation of gefitinib for 1 month. Thereafter, for all AEs that occurred, unless the investigator determined that remission was unlikely due to the patient's underlying disease, all AEs that persisted must be followed until resolution.
2) Evaluation of curative effect: The evaluation method of brain is MRI, and the evaluation method of primary tumor is CT, and the evaluation is carried out every 8 weeks. Evaluation of other extracranial tumors depends on the specific situation of different patients, and the time and method are determined by the investigator.
3) Survival follow-up: every 4 months until the patient died.

Primary endpoint
Comparison of intracranial progression-free survival (iPFS: defined as time from randomization to intracranial progressive disease or death.) in gefitinib plus chemotherapy group and gefitinib alone group.

Secondary endpoints
(1). Comparison of progression-free survival (PFS: time from randomization to overall disease progression or death) in gefitinib plus chemotherapy group and gefitinib alone group.
(2). Comparison of overall survival (OS: time from randomization to death from any cause) in gefitinib plus chemotherapy group and gefitinib alone group.

Sample Size
The sample size calculation of this study is based on the assumption that, for patients with EGFR-mutant NSCLC with brain metastases, gefitinib combined with pemetrexed/platinum therapy can prolong the intracranial disease progression-free time of patients compared with gefitinib alone. Based on iPFS reported in the literature, we determined that gefitinib monotherapy provided an iPFS of 8 months, the HR of gefitinib plus pemetrexed/platinum was 0.67, and the mean iPFS was 12 months. The analysis had 80% power to detect differences at the one-sided 10% significance level.
With 80 patients in each group, a cumulative enrollment period of 12 months, and a follow-up period of 12 months. Based on these assumptions, calculations were performed using the PASS 11 software.

Statistical analysis
The two-sided alpha value was 0.05 for all examinations performed, unless otherwise stated. All interaction checks that will be performed will have a two-sided alpha value of 0.1, and all CIs refer to the two-sided 95% value unless otherwise stated.
The primary endpoint was iPFS. The primary analysis compared iPFS and PFS in the groups receiving gefitinib in combination with pemetrexed/platinum and those receiving gefitinib monotherapy, with a one-sided significance level of 0.10. Cox regression models will be applied to estimate differences between treatment groups adjusted for selected prognostic factors (for stratification, age, smoking history, mutation type, presence or absence of symptoms of intracranial hypertension).
The overall response rate (ORR), defined as the proportion of tumor-response eligible patients with a PR or CR tumor response, will be calculated for each treatment group.
The disease control rate (DCR), which is the proportion of randomized patients whose tumor response was PR or CR or SD, will be calculated for each treatment group. ORR and DCR were also calculated as the proportion of patients in the tumor-eligible population undergoing sensitivity analysis. Fisher's exact test will be used for comparison of treatment groups.
The safety analysis of this study will include the total number and incidence (per treatment group) of laboratory and non-laboratory highest-grade CTCAE (version 4) adverse events, and their association with study drug. Toxicity rates will be compared between treatment groups.

Experimental group
Patients received gefitinib (250mg once daily) plus pemetrexed-platinum chemotherapy (pemetrexed 500mg/m 2 combined with cisplatin 75mg/m 2 or nedaplatin 80mg/m 2 ) in a 4-week cycle, for four to six cycles, followed by gefitinib plus pemetrexed maintenance. Patients continued treatment until intracranial progressive diseases, unacceptable adverse, or any cause of death.

Control group
Patients received gefitinib alone (250mg once daily) until intracranial progressive diseases, unacceptable adverse, or any cause of death.

Dosing schedule
Based on clinical data, pemetrexed 500 mg/m 2 was chosen to be administered on the first day of each cycle as an intravenous infusion over 10 minutes, every 28 days as a cycle.
Based on clinical data, cisplatin 25 mg/m 2 intravenous infusion daily on days 1 to 3 of each cycle, or nedaplatin 80 mg/m 2 was administered intravenously on the first day of each cycle, every 28 days as a cycle.
The dose of gefitinib is 250 mg orally daily in 28-day cycles.

Dose administration
A cycle is defined as an interval of 28 days (period delays due to holidays, weekends and bad weather and other unforeseen circumstances are permissible and not considered against the plan). Each cycle consisted of a combination of pemetrexed and platinum on the first day of treatment. Actual pemetrexed, cisplatin /nedaplatin doses should be calculated based on body surface area before the start of each cycle. A ± 5% difference is allowed between each administered dose and the calculated total dose.
The patients are required oral gefitinib 250 mg once daily. Tablets are taken on an empty stomach 1 hour before or 2 hours after a meal with 200 ml of water.

Pemetrexed-Related Toxicity, Dose Adjustment or Delay in Treatment
At the start of subsequent cycles, dose adjustments were made based on the hematologic counts or the maximal non-hematologic toxicity from the previous treatment cycle.
Treatment can be delayed to ensure adequate recovery time. Any patients requiring dose reduction continued to receive the reduced dose during the study. Any patient who undergoes two steps of dose reductions and develops toxicity requiring the third dose reduction must be withdrawn from pemetrexed/ platinum therapy. If patients dropped out of pemetrexed/ platinum therapy, they could continue to receive gefitinib rather than withdraw from the study. Treatment can be delayed up to 42 days from Day 1 of the current cycle to allow sufficient time for patients to recover from study drug-related toxicity. Patients who were unable to receive study drug within 42 days of their last treatment had to be withdrawn from pemetrexed/ platinum unless approved by the sponsor to continue the trial.

Hematological toxicity
At the start of subsequent cycles, dose adjustments were made based on the minimum of platelet counts and neutrophil counts from the previous treatment cycle. The absolute neutrophil count (ANC) must be ≥1.5 x 109/L and the platelet count must be ≥100 x 109/L prior to the start of any cycle. Blood tests should be performed for each patient within 7 days prior to randomization, within 4 days prior to dosing of each cycle (except for cycle 1, which can be performed within 7 days prior to randomization), and at the first follow-up visit after the end of treatment. school inspection. Treatment can be delayed to allow sufficient time for the patient to recover. Recovered patients must be re-treated according to the guidelines provided in

Nephrotoxicity
( 1 ) Dose adjustment of cisplatin Muscular hepatic clearance (Ccr) before treatment should be ≥ 60 ml/min, and should be used before the start of each treatment cycle.
Cockcroft-Gault formula for evaluation. (If the results calculated by the formula differ significantly from those estimated from collected urine, consider using the most accurate result). The interval for retesting of creatinine clearance is recommended weekly but must be based on the opinion of the investigator. If the patient's creatinine clearance does not return to the above level (CrCl>45 mL/min) within 42 days after the previous pemetrexed dose , the patient must discontinue chemotherapy .

Grade 3/4 nausea, vomiting
If the patient develops grade 3/4 nausea or vomiting, the dose of cisplatin should be reduced to 60 mg/m 2 in subsequent courses , or the dose of nedaplatin should be reduced to 75% of the original dose .

Ototoxicity
If the patient develops hearing loss, new tinnitus, or new severe high-frequency hearing on the hearing curve If strength is lost, cisplatin /nedaplatin should be discontinued . However, if there is clinical benefit, patients can continue to be treated with pemetrexed and gefitinib.、

Neurotoxicity
When neurotoxicity occurred, dose adjustments for pemetrexed/ platinum and gefitinib are shown in Table 4.5 . Cisplatin should be discontinued if grade 3 or 4 neurotoxicity occurs, but pemetrexed can be continued in patients with clinical benefit.

Clinically significant exudation
If patients had clinically significant pleural effusion or ascites prior to randomization, biologics could be given to control pleural effusion, but pleural infusion of chemotherapy drugs was not allowed. and through drainage or other surgical treatment uncontrolled, patients should be excluded from the study.
If a patient develops clinically significant pleural effusion or ascites (as determined by symptoms or clinical examination) during treatment, drainage of the effusion should be considered prior to administration. However, patients should be discontinued from the study if the investigator believes that the effusion is indicative of disease progression.

Insufficient folic acid or vitamin B12 supplementation
Folic acid should be taken for at least 3 days and an injection of vitamin B12 should be given within 7 days before the first dose of pemetrexed. When pemetrexed is given later, folic acid should be taken for at least 14 days in the 28 days before Day 1 of the next cycle before pemetrexed is given.

Dose Adjustment or Delay in Patients Receiving Gefitinib
If the patient develops intolerable diarrhea (sometimes associated with dehydration), adverse cutaneous drug reactions, or any adverse event deemed by the investigator to be attributable to gefitinib treatment, short-term (up to 14 days) interruption and then reinstating the patient on 250 mg daily may successfully treat these adverse events.
In the event of an acute onset or exacerbation of pulmonary symptoms (dyspnea, cough, fever), gefitinib treatment should be discontinued and these symptoms should be investigated immediately and appropriate treatment initiated. If interstitial lung disease is diagnosed, gefitinib should be discontinued and the patient given appropriate treatment.
Patients with new-onset ocular symptoms, such as pain, should be medically evaluated and given appropriate treatment, including discontinuation of gefitinib and removal of trichiasis if they develop. After symptoms and ocular changes resolve, then physician decide whether to resume 250 mg daily therapy.
No dose adjustment is required when patients differ in age, weight, sex, race, or renal status, or in the presence of moderate to severe hepatic impairment due to liver metastases.
Patients in group A (gefitinib combined with chemotherapy) can continue to receive chemotherapy if gefitinib therapy is discontinued.

Blinding
This is an open label clinical trial.

Concomitant therapy
Patients were allowed to receive comprehensive supportive care concurrently during the study. No other chemotherapy, immunotherapy, cancer stimulation veterinary therapy, surgery, or other investigational drugs. Pain that is not effectively controlled by systemic therapy or local analgesia of metastatic lesions, allowing the use of palliative radiotherapy to irradiate small areas. If any other specific anticancer therapy is required due to disease progression, study treatment must be discontinued early.
Particular attention should be paid to the following concomitant treatments:

Colony stimulating factor
The use and prevent use of granulocyte colony-stimulating factor (G-CSF),

erythropoietin (EPO), and thrombopoietin-stimulating drugs (TPO) in accordance with
CSCO's standardized management guidelines for tumor radiotherapy and chemotherapy-related myelosuppression are allowed.

NSAIDs
Patients taking NSAIDs or salicylates (excluding low-dose aspirin [≤1.3 g/day]), NSAIDs should be discontinued 2 days before, on the day of, and 2 days after receiving pemetrexed. If patients are taking long-half-life NSAIDs (eg, naproxen, piroxicam, diflunisal, or nerbulimone), they should be discontinued 5 days before, on the day of, and 2 days after receiving pemetrexed. The use of opioid analgesics is not limited.

Calcium folinate rescue therapy
Since folic acid and vitamin B12 supplementation significantly reduced the number of episodes of grade 4 hematologic and grade 3/4 nonhematologic toxicities associated with pemetrexed treatment, the use of calcium folinate as a rescue agent is not expected.
However, this section provides information for when rescue medication is needed.

Treatment of diarrhea
Patients with diarrhea should be given standard antidiarrheal therapy according to local routine practice. Patients with CTCAE grade 3 or 4 diarrhea should be given intravenous fluids, octreotide, and antidiarrheals. If diarrhea is severe (requiring intravenous fluids) or associated with fever or grade 3 or 4 neutropenia, broad-spectrum antibiotics should be given. Patients with severe diarrhea and severe nausea or vomiting must be hospitalized, given intravenous fluids, and corrected for electrolyte imbalances.

Treatment of Febrile Neutropenia
Patients with febrile neutropenia with or without diarrhea should be seen in the hospital according to standard management procedures, and intravenous antimicrobial therapy should be initiated urgently.

Concomitant drugs and gefitinib
Gefitinib may interact with CYP3A4 inhibitors and CYP3A4 promoters; therefore, the study protocol does not allow the use of CYP3A4 inducers or inhibitors in therapy.
Drugs that induce CYP3A4, such as rifampicin or phenytoin, can reduce the efficacy of gefitinib by increasing its metabolism and reducing its plasma concentration. With regard to gefitinib, patient compliance with study medication will be assessed at each visit. By comparison retrieved and dispensed tablets to assess adherence.
Substantially poor compliance has been demonstrated (one visit for patients taking <80% or >120% of the intended study drug at intervals, the study center must inform the patient of the importance of study drug adherence and drug counts. Patients who continue to fall outside compliance will be discontinued treatment.

Efficacy assessment
The purpose of this study was to evaluate the safety and efficacy of gefitinib plus progression. Imaging progression was determined using the judgment of an independent data review committee. Patients who were alive and did not progress by the date of analysis will be cut off on the date of their last imaging assessment.

Secondary efficacy endpoints
Progression-free survival (PFS): from the date of randomization to the first observation of intracranial or extracranial disease progression (intracranial based on imaging or intracranial hypertension, whichever is extracranial imaging) If patients died from other causes before disease progression, the number of days from randomization to death was calculated.
Intracranial Objective Response Rate (iORR): The proportion of patients who achieved a complete or partial response with the best intracranial response .
Overall objective response rate (ORR): The proportion of patients with a complete or partial response to the overall best response.
Overall survival (OS): Time from randomization to death from any cause. Patients still alive at the time of analysis will have the date of their last contact as the cutoff date.

Cognitive function: Hopkins Vocabulary Learning Test-Revised (Hopkins Vethal
Leaming Test-Revise, HVLT-R) and Trail Marking Test (TMT) were used to evaluate.

Safety Assessment
All patients who received at least one study drug treatment will be included as the safety population for safety analysis. The patient's physical examination, vital signs, adverse events, and laboratory abnormalities were summarized. all adverse events should be reported and graded according to the NCI Terminology Criteria for Common Adverse Events (CTCAE) version 4.0.

Safety assessment (1) Vital signs and physical examination
Patients will undergo a comprehensive physical examination, including ECOG score, physical examination, height, weight, and measurements of vital signs such as heart rate, blood pressure, temperature, and respiratory rate.

(2) Laboratory examination
Blood routine, blood biochemistry, urine routine and screening pregnancy tests will be performed in the clinical laboratories of each research center.
This study will evaluate the following laboratory metrics.

Adverse Events
Subjects must be closely monitored for adverse events. Such monitoring includes

Neutrophil Count (ANC)
White Blood Cell Count (WBC) Hemoglobin (HGB) Platelet count (PLT) Red blood cell count (RBC) clinical laboratory examinations. Adverse events should be assessed in terms of severity, severity, and relationship to the test drug.
The investigator is responsible for evaluating the relationship between all adverse events and the study drug. However, the Principal Investigator may entrust the judgment of other qualified clinicians participating in this study, but remains responsible. Investigators must provide a list of qualified and commissioned personnel.

(1) Definition of adverse events
An adverse event is an unforeseen medical condition or deterioration of a pre-existing medical condition that occurs during or after the use of a drug, whether related to the study drug or not. Unforeseen medical conditions may be symptoms (eg, nausea, chest pain), signs (eg, tachycardia, enlarged liver), or abnormal test results (eg, laboratory tests, electrocardiogram). In clinical research, from the signed informed consent form, adverse events can be unforeseen adverse medical conditions that occur at any time, including screening or washout periods, even if have not received study treatment.
Adverse events in humans (whether drug-related or not) include the following: • Adverse events that occur during the use of the drug by professionals; • Adverse events resulting from drug overdose (whether intentional or unintentional); • Adverse events caused by drug abuse; • Adverse events caused by discontinuation of the drug; • Adverse events that may arise purely from patient participation in the study (eg, due to discontinuation of antihypertensive drugs during the washout period or serious adverse events) must be reported as adverse events even if not related to the study medication.
Those who did not appear or did not achieve the expected clinical pharmacological effects and have been recorded in the corresponding section of the CRF are not regarded as adverse events. However, if the criteria for "serious" adverse events are met, they should also be recorded and reported as serious adverse events. In this study, any event clearly caused by disease progression was not reported as an adverse event.
(2) Unexpected adverse events Unexpected adverse events are defined as any characteristic or severity consistent with the Investigator's Brochure (or instructions) inconsistent adverse drug reactions.
Characteristics or severity of known, documented adverse events with important information on the extent of the unexpected adverse event is also part of the reporting.
For example: (a) acute renal failure that has been identified as an adverse event followed by interstitial nephritis; and (b) first report hepatitis with acute hepatic necrosis.

(3) Observation, recording and reporting of adverse events
All adverse events that occurred after the subject signed the informed consent form must be completely recorded in the subject's case report form. Documentation must be supported by original sources. Each event should be described in detail, including the date of beginning and discontinuation, severity, relationship to the test product, actions taken, and outcome of the event.
① How to detect adverse events At each visit, adverse events can be detected by: • Unsolicited information from patients or caregivers; • At each visit, ask the patient open, non-leading questions: How are you feeling?
Have you had any (other) medical problems since your last visit?
• Abnormalities observed by investigators, other medical staff, and family members ② Time to collect adverse events In this study, non-serious adverse events were recorded from the time the patient gave informed consent until the patient 30-day follow-up period after withdrawal from treatment.

③ Collection of adverse event data
All adverse events should be recorded on the CRF. The description of the adverse event includes its start and end time, whether it is in line with serious adverse events, measures taken (such as changes in study treatment, other treatments and follow-up examinations) and outcomes, and the investigator is asked to evaluate the causal relationship (with the study treatment). Relationship). Adverse events should be graded by NCI CTC and their changes recorded on the corresponding CRF.

(4) Criteria for judging the severity of adverse events
The severity of adverse events should be recorded according to the National Cancer Institute-Common Toxicity Criteria Version 4.0 (NCI-CTCAE 4.0).

(5) Judgment criteria for the relationship between adverse events and trial drugs
Clinicians should assess the probability that an AE is related to experimental drug therapy and classify the probability into 5 categories 'definitely related, probably related, definitely unrelated, probably unrelated, unclassified'. Of which 'definitely related', 'probably related' and 'unclassified' are all seen as AE. The incidence of AE is calculated as the sum of AE cases ('definitely related' + 'probably related' + 'to be determined') divided by number of patients evaluable for safety. The criteria for causality assessment is shown in table 4.4. Table 4.4. Criteria for assessment of causal relationship between AE and experimental drug

Classification
Criteria for assessment definitely related The lag time prior to onset of the event is consistent with a reasonable chronological order after the last dose of the drug, and the event fits the characteristics of known reaction to the drug or other drugs of the same class; the severity of the event has reduced on drug dose interruption, and the manifestations of the event reappear after the drug is restarted.
probably related The lag time prior to onset of the event is consistent with a reasonable chronological order after the last dose of the drug, but the event does not fit the characteristics of known reaction to the drug or other drugs of the same class; it seems possible that the underlying clinical condition of the patient or one of the concomitant medications known to cause the event actually has caused it.
probably unrelated The lag time prior to onset of the event is not consistent with a reasonable chronological order after the last dose of the drug, and the event does not fit the characteristics of known reaction to the drug or other drugs of the same class; it seems possible that the underlying clinical condition of the patient or one of the concomitant medications known to cause the event actually has caused it.
definitely unrelated The lag time prior to onset of the event is not consistent with a reasonable chronological order after the last dose of the drug, and the event does not fit the characteristics of known reaction to the drug or other drugs of the same class; it seems possible that the underlying clinical condition of the patient or one of the concomitant medications known to cause the AE actually has caused it; the manifestations of the event disappear on the disease amelioration or concomitant medications interruption, and reappear after concomitant medications are restarted.

unclassified
The lag time prior to onset of the event is ambiguously consistent with a reasonable chronological order after the last dose of the drug, but the event fits the characteristics of known reaction to the drug or other drugs of the same class; it seems possible that one of the concomitant medications known to cause the event actually has caused it.
Five criteria for adverse reaction/event analysis: to the normal range and/or is sufficient to explain the abnormality. Record clear explanations on the case report form.

(6) Treatment, follow-up and duration of adverse event cases
Close follow-up of the patient after the investigator's initial report of an adverse event is required to provide further information to the study sponsor. All adverse events during the study, even if the patient has completed the study or ended treatment, should be followed up until remission or until the patient's condition is stable, unless according to the investigator's judgment, the adverse event is unlikely to be alleviated due to the patient's own condition, or patients were lost to follow-up.

Serious Adverse Events
(SAEs) were collected from the time patients signed informed consent and received study drug . If a patient developed an SAE after giving informed consent but before receiving study drug, this event will not be collected unless the investigator believes that the event was caused by a protocol procedure.
Surgery that was previously planned (before signing the ICF) should not be reported as an SAE unless the underlying disease worsened during the course of the study.
In the event of any serious adverse event (SAE), site personnel must report the event to the study sponsor within 24 hours of becoming aware of the event. Immediately after reporting by telephone, a study-specific SAE form should be completed as a formal notification. This 24-hour notification requirement refers to the initial initial SAE information and all follow-up follow-up SAE information. A SAE is any AE in a study that results in one of the following: • Death, except due to disease progression (see "Exclusions" below); • lead to or prolong the hospital stay; • life-threatening conditions (risk of immediate death); • Persistent or significant disability/disability; • Congenital anomalies/birth defects; • Other reasons judged by the investigator to be serious Medical events that are important, but do not result in death, life-threatening disease, or require hospitalization, if, using appropriate medical judgment, they are likely to endanger the patient and require medical and surgical intervention to prevent any of the outcomes listed in this definition, then This event can be considered a serious adverse drug event.
Exclusions: Death due to disease progression will only be reported to the study sponsor as an SAE if the investigator believes it is related to the use of the study drug.
Serious adverse events that occurred within 30 days of the patient's last dose were collected, regardless of the investigator's view of causality. Thereafter, reporting of SAEs is not required unless the investigator believes the event is related to the study drug, or the drug delivery system, or the operation of the protocol.
Information on SAEs in the study population that are not expected to be related to drug exposure, and information on SAEs that will be reviewed periodically by the sponsor during the course of the clinical trial. In a population with advanced non-small cell lung cancer, the incidence of serious infections, cardiovascular events, and weight loss can be reasonably predicted based on a variety of factors, such as age, comorbidities, and disease status.

Suspected Unexpected Serious Adverse Reactions
Suspected Unexpected Serious Adverse Reactions (SUSAR) were defined as serious, not listed in the Investigator's Brochure, investigators identify events that are believed to be related to the study drug or procedure.

Other security indicators
Other safety measures include physical examination (to identify rash or third space fluid accumulation, and possible signs suggestive of AE (such as body temperature, weight, blood pressure, and heart rate) and toxicity grade.
The National Cancer Institute (NCI)-CTCAE v4.0 was used as a reference when selecting appropriate terms and grading severity. When an AE had no matching term in the NCI-CTCAE v 4.0 criteria, the investigator was responsible for selecting the appropriate system organ class and rating the severity of the event based on the intensity of the event. Note that both the CTCAE term (actual or numbered) and severity must be selected by site personnel and recorded on the case report form (CRF). This record is a supplement to the text describing the AE verbatim. Before each cycle, patients will be graded for toxicity (applying the NCI-CTCAE v4.0 criteria).

Drug Safety Monitoring
Throughout the course of the study, the clinical study physician will monitor safety data.

Declaration
This clinical trial will be conducted in strict compliance with protocol, GCP and regulatory and SOP requirements.

Ethics
It is the investigator's responsibility to provide the ethics committee with the clinical trial protocol, informed consent, and information to obtain an independent approval document for the conduct of the clinical study.
The approval document of the ethics committee must be obtained before the start of the clinical study. Approval from the ethics committee must be in writing and sent to the investigator, who will then provide a copy of the approval to the study sponsor. The ethics committee approval document must be accompanied by a list of all committee members involved in the discussion of the approval document and their respective responsibilities.
During the course of the clinical study, any issues related to the safety of the clinical study, such as the clinical study protocol or changes to subjects' informed consent and serious adverse events in clinical research must be reported to the ethics committee.
The end or early termination of a clinical study must also be reported to the ethics committee.

Raw data verification
For data recorded directly on the case report form (i.e. data not previously recorded in writing or electronically) with and the identification of the raw data considered as the original data should be stipulated and clearly stated in the monitoring plan in advance according to the plan. Otherwise, it is regarded as lack of raw data.
Investigators must properly handle all data obtained during clinical research to ensure the rights and privacy of subjects in clinical trial. The investigator must agree with the monitor/auditor/inspector on the required clinical research data are reviewed and reviewed in order to verify the accuracy of the original data and understand the progress of the research. If the original records cannot be verified, the investigator should agree to assist the monitor/auditor/inspector further confirm the quality of the data.

Quality Control and Assurance
All drugs and materials used in clinical research must be subject to quality control.
sponsor, sponsor persons authorized by the party or relevant medical management institutions have the right to conduct audits of clinical research, the purpose is to ensure the authenticity of clinical research record data and comply with the provisions of the clinical research protocol. clinical research subjects will be informed that there will be relevant personnel to audit during the trial, but patient privacy and data will be strictly protected.

Informed Consent/Data Protection Agreement
The investigator is responsible for explaining the purpose, methods, benefits and potential risks of this clinical trial to each subject, and obtaining an informed consent form signed by the subjects of the clinical trial. Before the start of the procedure, the subject's informed consent must be obtained. For subjects who can not signed the informed consent, the informed consent must be signed by their parents, legal guardians or guardians. By signing an informed consent form, subjects must also consent to the monitor/auditor/health survey organization checks the raw data obtained about the clinical study in order to determine the reliability of clinical study data results.
The original informed consent form signed and dated by the subjects of the clinical trial must be properly kept by the investigator , and the signed informed consent form must also be recorded in the case report form and related original records of the trial.

Modification of clinical protocol
The modification of the clinical protocol requires the joint participation of the sponsor and the investigator. A revised version of the protocol should be submitted to the ethics committee will be approved. Only then can the contents of the revised version be executed.

Case report form
The investigator must ensure that the electronic case report form is completed and accurately completed. Only record in each case report form record data of a clinical study subject.

Subject Privacy
Researchers must ensure that the privacy of clinical trial subjects is maintained. In all documents submitted to the sponsor, the identity of the clinical trial subjects can only be determined by the trial patient number and initials, and cannot be specified subject's full name. Investigators must properly keep the names and addresses of clinical trial subjects and the corresponding entry form. These enrollment forms are kept strictly confidential by the investigator and cannot be submitted to the sponsor.

Publishing papers
For multicenter clinical studies, data from each center cannot be individually published.
The sponsor reserves the right to review the first draft prior to publication or publication of the study results.

Data archiving
The investigators should retain all relevant records for a period of at least 5 years after completion of the study in compliance with Chinese GCP requirements. b. Premenopausal women of childbearing potential must have a negative urine or serum pregnancy test within 7 days before the first dose. If the result is positive, the subject will not be eligible to participate in the study. If pregnancy is suspected during the test, the test should be repeated.

Study Flowchart
c. If assessments were performed within 7 days before randomization and the listed inclusion and exclusion criteria (if applicable) were met, repeats were not required on the day of the first dose unless the investigator believed that significant changes were likely.
d. Physical examination includes heart rate, blood pressure, respiratory rate, temperature, height, weight, and neurological examination. If body weight was measured at screening, repeating is not necessary unless clinically indicated. Body weight should be measured at screening and when clinically indicated.
e. Data on concomitant medications, including drug dose, route of administration, dosing schedule, start date, indication, end date, and end reason, were collected and must be recorded from randomization with study drug to 1 month after discontinuation. Treating physicians obtained this information from patients' hospital visits every 6 weeks. With knowledge of the patient's routinely prescribed medications, the treating physician will assess whether changes in the patient's concomitant medications over the past 8 weeks are likely to lead to clinically meaningful improvement in pulmonary symptoms.
f. The screening period blood routine/blood biochemistry/urine routine examination should be performed within 7 days before randomization, and the subsequent blood routine, blood biochemistry and urine routine examination should be performed within 7 days before and after the time point specified in the study flow chart. Routine blood tests were performed every 3 days in each phase of chemotherapy, and blood biochemical tests were performed every four weeks and when clinically indicated.
g. Enhanced magnetic resonance imaging for intracranial lesions and enhanced computerized tomography scans for extracranial lesions. Each investigator decides which imaging modality to use, but screening assessments must adequately define aspects of the disease and document measurable lesions according to RECIST  i. In the event of any perceived ocular symptoms, new or worsening respiratory symptoms (eg, cough, wheezing), the patient should be given immediate appropriate medical attention. Any symptoms were handled as clinically routine and, if defined, were reported as adverse events or serious adverse events (SAEs). Adverse events were collected from patients signing informed consent. After study termination, all unhealed or serious adverse events were followed until resolution, unless the investigator determined that remission was unlikely due to the patient's own disease.
j. After recording disease progression, investigators were required to contact the patient, the patient's family, or the patient's current treating physician by telephone at least once every 4 months to obtain information on the patient's overall survival and post-study chemotherapy until the patient's death.
The following information should be obtained during each follow-up visit: disease status, records of all new anticancer treatments, and date of death (if applicable).