The introduction of modern staging systems has increased the detection of small peripheral lung cancers at an early stage [1]. Stage I non-small-cell lung cancers (NSCLCs) are confined to the lung without lymph node involvement, and surgical resection is currently considered the standard therapeutic approach.

Nodal staging is initially performed non-invasively with computer tomography (CT) and positron emission tomography (PET) scans followed by minimally invasive staging with endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) when CT and/or PET are suggestive of mediastinal nodal involvement. Lobectomy with radical lymphadenectomy is currently considered the treatment of choice for early-stage lung cancer.

Several studies demonstrated that primary invasive non- small-cell lung carcinomas > 2.0 cm were twice as likely to have nodal metastases as carcinomas ≤ 2.0 cm, emphasizing that small lung cancers had less lymph node involvement and confirming a better survival. In our pilot study [18] published in 2011 in the European Journal of Thoracic Surgery, no nodal involvement was observed in any of the 62 patients with pulmonary nodule size less than 10 mm, in 20 out of 120 patients (17%) with nodule size 11-20 mm, and in 9 out of 37 tumors (24%) 21-30 mm in size (p = 0.0007).

These patients could be spared radical lymph node dissection if deemed not essential, thereby reducing operative risks, postoperative morbidity, and surgery time. A preoperative diagnostic determination to establish the size and correct staging of the tumor is mandatory for appropriate selection of candidates, avoiding unnecessary surgery.

BACKGROUND AND RATIONALE The introduction of modern staging systems such as computed tomography (CT) and positron emission tomography/CT (PET/CT) with fluorodeoxyglucose (FDG) has increased the detection of small peripheral lung cancers at an early stage [1]. Stage I non-small-cell lung cancers (NSCLCs) are confined to the lung without lymph node involvement, and surgical resection is currently considered the standard therapeutic approach. Lobectomy with radical lymphadenectomy is currently considered the treatment of choice for early-stage lung cancer, irrespective of tumor size or its metabolic features on PET. However, the new TNM classification recently demonstrated that very small lung cancers may be less aggressive than others [2], suggesting a less aggressive surgical approach to reduce morbidity.

According to the new TNM classification (8th edition), the overall 5-year survival rate for pathological stage IA NSCLC was 85%, ranging from 80% to 90% for T1a and T1c tumors respectively, compared with 64% for stage IB tumors [2]. Among patients with stage I cancer, tumor size may affect outcome and drive survival, as confirmed by different studies [7-12].

Several studies demonstrated that primary invasive non- small-cell lung carcinomas > 2.0 cm were twice as likely to have nodal metastases as carcinomas ≤ 2.0 cm, emphasizing that small lung cancers had less lymph node involvement, and confirming a better survival [11-15]. Based on that, Ishida et al. and then Konaka et al. have already demonstrated the absence of lymph node involvement in sub-centimeter lung cancers and the feasibility to omit lymph node dissection in those cases [15,16]. However, Zhoua et al. recommended systematic nodal dissection in the presence of sub-centimeter disease, finding nodal metastases in 15% of NSCLC < 1 cm, but also including higher tumor stages such as stage II and III in their study without performing PET scan as part of the preoperative staging [17]. In our pilot study [18] published in 2011 in the European Journal of Thoracic Surgery, no nodal involvement was observed in any of the 62 patients with pulmonary nodule size less than 10 mm, in 20 out of 120 patients (17%) with nodule size 11-20 mm, and in 9 out of 37 tumors (24%) 21-30 mm in size (p = 0.0007). All 55 patients with nodule SUV < 2.0 and all 26 non-solid lesions were pN0 (respectively, p = 0.0001 and p = 0.03). So we adopted cut-offs of 10 mm for nodule diameter and 2.0 as peak SUV to distinguish patients with nodal involvement from those without.

These patients could be spared radical lymph node dissection if deemed not essential, thereby reducing operative risks, postoperative morbidity, and surgery time. A preoperative diagnostic determination to establish the size and correct staging of the tumor is mandatory for appropriate selection of candidates, avoiding unnecessary surgery.

Data have demonstrated that positron emission tomography (PET-FDG) is a reliable tool versus CT in evaluating both solitary pulmonary nodules and lymph node involvement [3]. Standard nodal staging is initially performed non-invasively with PET followed by minimally invasive staging with endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) when PET is suggestive of mediastinal nodal involvement. In fact, EBUS-TBNA is a minimally invasive procedure with a high yield for lymph node staging of lung cancer [4]. EBUS-TBNA allows access to the paratracheal lymph node stations (levels 2R, 2L, 4R, 4L), the subcarinal lymph node (level 7), hilar, interlobar, and lobar lymph nodes (levels 10, 11, and 12). Previous studies, including systematic reviews and meta-analyses, and more recently Yasufuku et al. have demonstrated a major impact of EBUS-TBNA on management of patients with non-small cell lung cancer (NSCLC), with a diagnostic yield comparable to mediastinoscopy [5,6]. Annema et al. have recently confirmed that the sensitivity of endosonography is similar to that of mediastinoscopy (85% vs 79%, respectively), and associated with a lower complication rate (1% vs 6% for mediastinoscopy), so concluding that endosonography should be the first step for mediastinal nodal staging [19].

In patients without evidence of mediastinal nodal metastasis on PET and CT, the need for EBUS-TBNA becomes less clear. The most recent American College of Chest Physicians (ACCP) evidence-based guidelines recommend minimally invasive mediastinal staging with a needle technique in patients with a central tumor or nodal hilar disease (N1) on PET/CT. This is based on the increased prevalence of N2 disease in this group, but this is a grade 1C recommendation, that is based on low quality evidence. In patients with peripheral tumors and no evidence of mediastinal or hilar nodal disease on PET/CT, invasive staging is not recommended.

Concerns have been raised regarding these recommendations given that the prevalence of occult nodal metastasis in patients with N0 disease by PET/CT appears to be higher than previously reported, with recent studies showing values as high as 17-22% [20,21]. EBUS-TBNA may provide an attractive option to increase staging accuracy. However, sensitivity of EBUS-TBNA in patients with clinical N0/N1 disease on PET/CT is unclear and largely based on retrospective studies [22,23]. The only prospective study published by Leong et al. [24] demonstrated that a significant proportion of patients with N0/N1 disease by PET/CT had N2 disease (20%) and EBUS-TBNA was able to identify a substantial fraction of these patients, thus improving diagnostic accuracy compared with PET/CT alone. Sensitivity of EBUS-TBNA however appears lower compared with historical data from patients with larger volume mediastinal disease [5,6,26].

TRIAL OBJECTIVES

Hypotheses:

• In selected series the overall sensitivity of EBUS varies from 80-90% (88%) but for clinical N0/N1 drops to 40-49%.
• About 24% of patients with early stage lung cancer > 2 cm and less than 5 cm (T1b,c and T2a,b N0 PET-) would be N+ , whereas none of the T1a (less or equal than 1 cm) NSCLC would be upstaged considering the absence of N+ [19].

From reference 19 pT size ALL pN+ <10 mm 62 0 ( 0%) 11-20 mm 120 20 (17%) >20 mm 37 9 (24%)

7. OVERALL DESIGN AND PLAN OF TRIAL

7.1 Design of the study All patients clinical stage I and II will be undergone EBUS-TBNA for the staging of hilar and mediastinal lymph nodes.

All patients will be undergoing robotic (RATS) or videothoracoscopic (VATS) lobectomy plus radical lymph node dissection following the Society of Thoracic Surgery (STS) guideline. All lymph nodes previously sampled with EBUS-TBNA will be compared with those harvested during surgery.

Statistical analysis Observed (Surgery) pN- pN+ Predicted EBUS Negative TN FN UPSTAGING = (FN+TP)/total patients Positive FP TP SENSITIVITY is defined as the ration of TP/ (TP+FN) A sample size of 12 upstaged patients will achieve 80% power to detect a difference of sensitivity >38.4% assuming that the sensitivity under the null hypothesis is 49% (results of the meta-analysis by Leong et al, 2018 [22]), and that the actual sensitivity is >88% (95.6% in the study by Guarize et al, 2018 [25]) using a two-sided binomial test.

N° Upstaging EBUS Upstaging Surgery Sensitivity Leong 2017 Meta-analysis 15% (6- 24%) n/a 49% Naur 2017 167 6% (10/167) 10%(12/115) 43% (10/21) Ong 2015 220 8% (18/220) 27%(27/100) 36.7% * Shingyoji 2014 113 6% (7/113)** 17.6% (20/113) 35% (7/20) Vial 2018 75 8% (6/75) 20% (15/75) 40% (6/15)

Assuming that the proportion of patients with pathological pN+ (upstaging) is 24%, a total sample of 50 patients will be necessary to be enrolled in the study.

This hypothesis will be rejected if 2 or more FN are observed among the 12 upstaged patients.

ETHICAL CONSIDERATIONS

Patient protection The responsible investigator will ensure that this study is conducted in agreement with the Declaration of Helsinki.

The protocol will be approved by the Local Ethics Committee.

Subject identification A sequential identification number will be automatically attributed to each subject/patient registered in the trial. This number will identify the subject/patient and must be included on all case report forms.

Informed consent The investigator must provide to each subject/patient both oral and written information about the Trial and must ensure that the subject is fully informed about the aims of the trial, procedures, potential risks, any discomforts and expected benefits.

The subject/patient must agree that his/her data will be processed and stored in an anonymous form for evaluation of this trial and any later overviews and that his /her data may also be transferred in an anonymous form to third parties.

It must be emphasized that participation is voluntary and that the subject has the right to withdraw from the Trial at any time without prejudice.

A physician must obtain the subject's voluntary, personally signed (and dated) Informed Consent prior to any Trial - related procedure.

DATA SUBMISSION We will conduct the trial according to the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) Good Clinical Practice (GCP) guidelines.

Case report forms Data will be collected as for trial specific case report forms.

Signing and submitting forms All forms should be signed by the Principal Investigator or designee.

Data management Data collected in this trial will be sent to the Data Management Office of the European Institute of Oncology. The Data Management Center will process the data and will generate queries and forms requests. The statistician will perform the data analysis.

Authorization log The Principal Investigator (PI) should identify the other members of the Clinical Trial Team who are supervised by the PI and approved to provide information in case report form (CRF), queries, etc.

Subject/Patient identification log As per GCP, subjects/patients have the right to confidentiality. Therefore, no subject'/patients' names should be used in CRFs or any other documentation transmitted to the data management. Item that are used to identify a subject/patient include initials of subjects/patient's name, date of birth, registration number.

ADMINISTRATION RESPONSABILITIES The Chairmen of the study will be responsible for writing the protocol, reviewing all case report forms and documenting their review on evaluation forms, the contents of the reports, and for publishing the study results. They will also generally be responsible for answering all clinical questions concerning eligibility, treatment and the evaluation of the subjects/patients.

PROPERTY OF DATA AND PUBLICATION POLICY Property of data is of European Institute of Oncology. The main results of the clinical trial will be published in a peer-reviewed scientific journal. The final publication will be written by the Study Chairman or by one or more co-investigators of the study.

All publications, abstracts or presentations including data related to the present trial will be submitted for review to the Chair of the study prior to submission.

Trials results will not be released before data maturity has been reached for the primary endpoint of the trial.

Inclusion Criteria:

• Suspected or proven non-small cell lung cancer (NSCLC) clinical stage I and II (diameter > 1 cm and less than 5 cm, no pleura invasion) clinical N0M0 (8th TNM)
• All patients have to be staged by total body CT scan and PET-FDG
• Negative preoperative staging at hilar and mediastinal level at CT and CT/PET (PET negative and lymph node short axis < 1 cm at CT scan)
• Age between 18 and 75 years old

Exclusion Criteria:

• NSCLC smaller or equal than 1 cm
• Unfit for bronchoscopy or surgical resection
• Evidence of locally advanced or metastatic disease
• Prior chemotherapy or radiotherapy for this malignancy
• Other malignancy within the past 5 years except for not melanoma skin cancer, superficial bladder cancer or carcinoma in situ of the cervix
• Previous surgical treatment for lung cancer
• Multiple lung tumors