Accurate preoperative mediastinal clinical staging and post-treatment restaging of resectable non-small cell lung cancer is essential for treatment decisions. In 2007, the European Society of Thoracic Surgeons (ESTS) published a guideline for preoperative mediastinal staging that integrates imaging, endoscopic, and surgical techniques. In the last few years, several mediastinal staging techniques have emerged and more evidence has been published.
Therefore, a new version of the ESTS guidelines is needed: histologic confirmation is recommended for any mediastinal lymph node enlargement suggested by CT or PET-CT. Ultrasound endoscopic fine-needle aspiration biopsy (EBUS-TBNA) is the method of choice when available because it is not only a minimally invasive technique, but also has a higher sensitivity to exclude mediastinal lymph node metastases. If EBUS-TBNA is negative, then surgical techniques can be staged along with lymph node dissection or biopsy.
And the recommended level of TV mediastinoscopy is higher than that of plain mediastinoscopy. The highest accuracy can be obtained by combining endoscopic staging techniques with surgical techniques. Assuming no enlarged lymph nodes are seen on CT or no high uptake lymph nodes are seen on PET-CT, surgical resection plus systemic lymph node dissection can be performed directly for peripheral lung nodes ≤3 cm.
For patients with central tumors or with N1 lymph node involvement, preoperative mediastinal staging is routinely recommended. Whether to perform EBUS/EUS (transesophageal ultrasound) fine-needle aspiration or mediastinoscopy is up to the discretion of the local physician, but the principle is to use the least invasive examination. Preoperative mediastinal evaluation is also routinely recommended for lesions larger than 3 cm, especially in patients with adenocarcinoma with high SUV uptake.
Background
In patients with non-small cell lung cancer without distant metastases, mediastinal staging is important to make an accurate determination of the extent of disease, to guide appropriate treatment modalities, and to determine patient prognosis.
In 2007, ESSS published guidelines for preoperative mediastinal staging based on the evidence at that time. This guideline integrates imaging, endoscopic techniques, and surgical techniques. This guideline is very broad in application and very good in practice. It has a negative predictive value of 0.94.
New techniques have led to a better understanding of mediastinal staging. Therefore, the ESTS committee agreed to the work of the working group to undertake the revision and update of the old guideline.
For the elaboration of preoperative mediastinal lymph node staging
Current guidelines for the treatment of lung cancer are determined by the clinical status of the mediastinal lymph nodes. The purpose of mediastinal staging is to exclude patients who are likely to have mediastinal lymph node metastases; after all, these patients do not benefit from surgery.
The optimal treatment of N2 disease is currently controversial and is determined by the heterogeneity of the lymph nodes. All patient characteristics, tumor features, and extent of resection can influence the choice of treatment modality.
Patients with pathologically confirmed N2 involvement prior to treatment require comprehensive surgical multidisciplinary treatment, and these patients are first treated with induction chemotherapy or induction radiotherapy. Assuming that mediastinal lymph nodes achieve a step-down or that remission of these lymph nodes and tumor is significant, surgical resection and systemic lymph node dissection can be performed, resulting in an appreciable five-year survival rate with acceptable complications.
A number of prognostic predictors are currently recognized, some of which are related to the primary tumor and others to the extent of lymph node disease. A patient entering surgical multidisciplinary comprehensive care should first have a lesion that is technically resectable.
Patients who have unresectable lesions, such as extra-nodal metastases (clearly visualized by mediastinoscopy) or whole fused N2 involvement on CT, cannot proceed to the next surgical multidisciplinary treatment. These patients should be absolute indications for radiotherapy, assuming their physical condition allows it.
Whole N2 disease is difficult to define, but it has some relevance to the imaging “Group A lesions” described by the American College of Chest Surgeons (ACCP) evidence-based clinical practice guidelines. This group of imaging is defined as mediastinal gonadal infiltrative fusion, which is difficult to assess.
It is difficult to evaluate and distinguish individual lymph nodes. The concept of a mass is not strictly defined by the size of the lymph node, but according to ACCP guidelines, a lymph node with a short diameter of more than 25 mm is defined as a mass involvement (grade V).
Lump involvement can be manifested as single station or multi-station/multi-regional lymph node involvement. Since this article discusses preoperative lymph node staging, techniques for obtaining whole mediastinal lymph node disease will not be discussed here.
Preoperative mediastinal lymph node staging
Although we should aim for the highest sensitivity and negative predictive value, the working group also believes that a pathological N2 within 10% is acceptable after a rigorous evaluation. After thorough mediastinal staging, this missed pathologic N2 is also generally single-station and resectable.
There are several techniques available, but these depend on local medical practice and practice
These techniques include
1. Imaging techniques
1.1 Chest CT
Chest CT has always had an important place in lung cancer imaging. However, due to its low sensitivity (55%) and specificity (81%), it is not possible to choose how to perform biopsy of the tissue by relying on the anatomical images provided by CT alone.
1.2 PET-CT scan
PET-CT in combination with CT can further improve the accuracy of lymph node staging with an overall sensitivity of 80-90% and specificity of 85-95%. PET-CT has a high negative predictive value for detecting mediastinal lymph node disease in peripheral non-small cell lung cancer. Except for the following cases.
I. suspicious N1 lymph node metastasis
II. tumors >3 cm
III. Central tumor with no suspicious lymph node metastasis on CT or PET
In a Japanese study, 30% of the 143 patients with N1 lymph node involvement (lymph nodes >1 cm in short diameter) on CT had pathologically confirmed N2 or N3 involvement.
A recent meta-analysis confirmed that the negative predictive value of PET-CT was 94% (649 patients) in patients with tumors ≤3 cm and 89% (130 patients) in patients with T2 (6th edition TNM stage) with tumors >3 cm. This finding was confirmed in a recent prospective study by Spanish scholars: the predictive value of PET-CT was 92% for tumors ≤3 cm in the periphery, but only 85% for patients with >3 cm in the periphery.
Based on these studies, we now recommend that patients with peripheral type ≤3 cm tumors without hilar or mediastinal lymph node enlargement on CT and PET-CT may be excluded from further mediastinal staging. The rate of mediastinal lymph node metastasis was found to be higher in adenocarcinoma than in other tumors (risk ratio 2.72). Also high FDG uptake in the primary lesion increases the risk of unintended mediastinal lymph node metastasis.
For patients with tumors >3 cm (mainly adenocarcinoma with high FDG uptake), further pathology should be performed to exclude mediastinal metastases.
Prof. Lee et al. examined the postoperative confirmation of pathologic N2 in patients with clinical stage I (no hilar or mediastinal enlarged lymph nodes by PET and CT) non-small cell lung cancer. 2.9% of stage I peripheral lung cancers had pathologic N2 compared to 21.6% of central lung cancers.
1.3 Magnetic resonance examination
Advances in magnetic resonance imaging have allowed us to obtain diffusion-weighted imaging nuclear magnetic images (DWI), which provide very good tissue contrast. This technique reflects a large amount of valid information at the cellular level and provides images of the cytoarchitecture and complete cell membrane structure of the tumor.
In
In a recent meta-analysis, the accuracy of both DWI and PET-CT was evaluated and the sensitivity of DWI for integration was 0.95, significantly higher than that of PET-CT at 0.89. However, there are no large prospective trials comparing DWI and PET-CT.
However, there are no large prospective trials comparing the advantages and disadvantages of DWI and PET-CT techniques, and it is not possible to assess the value of DWI techniques in non-small cell lung cancer at this time.
2. Endoscopic techniques
2.1 Conventional fine needle aspiration: Although the conventional TBNA technique has been used for nearly 30 years, however, only a small percentage (10-15%) of patients with previously resectable stage I-III non-small cell lung cancer have undergone TBNA for mediastinal lymph node staging.
The main reasons limiting its use are the size of the nodes (CT short diameter >15-20 mm) and the technique used to perform it. A meta-analysis reported a sensitivity of 78% and a false-negative rate of 28% for the TBNA technique. Conventional
TBNA blind puncture is very much in place if it confirms N3 involvement, but endoscopists often do not perform further punctures to exclude N3 involvement once N2 involvement has been confirmed.
2.2 Ultrasound endoscopy: transesophageal ultrasound puncture and transtracheoscopic ultrasound puncture. Practical aspects: Although transesophageal (tracheoscopic) ultrasound fine needle puncture can be performed under general anesthesia in some centers, most centers perform it in outpatients with local anesthesia and light sedation.
EBUS is able to visualize upper and lower mediastinal lymph nodes, including 2R/2L, 4R/4L and station 7, as well as hilar lymph nodes at stations 10, 11 and even 12 according to the latest lymph node anatomy atlas.
Therefore, EUS can compensate for the lack of other means, such as lymph nodes at some stations (e.g., 8, 9, etc.) that are difficult to obtain with EBUS-TBNA or mediastinoscopy. Although some specialized centers believe that EUS can puncture lymph nodes at stations 5 and 6, the limited data currently available do not allow us to recommend this puncture.
What is achievable with current technology is that at least 5 mm of lymph nodes can be visualized as well as adequately sampled, and studies have found that the optimal number of punctures per station is 3. Assuming that the patient requires a mediastinal lymph node staging, the use of
endoscopic technique is fully achievable for systemic lymph node biopsy. Indeed, several ultrasound endoscopy studies have demonstrated a number of stations for mediastinal lymph node biopsy of 3-4 stations per patient. We stipulate that during ultrasound endoscopy groups 4R, 4L and 7 are mandatory biopsies and described in the report In addition to this, lymph nodes >5 mm detected by ultrasound, as well as lymph nodes with high FDG metabolism are subject to biopsy and pathological examination. Moreover, according to the guidelines, endoscopists are allowed to perform biopsies in the 10R and 10L groups. To avoid contamination and tumor dissemination by using one puncture needle, biopsy of lymph nodes should start with N3 lymph nodes, followed by N2 station lymph nodes and finally N1 station lymph nodes should be examined.
Performance characteristics: Several meta-analyses have shown that EUS puncture by itself, EBUS puncture by itself, and EUS combined
The sensitivity of EBUS puncture in mediastinal staging of lung cancer ranged from 83% to 94%. Only one randomized controlled trial (Aster trial) compared mediastinoscopy as specified in the 2007 edition of the ESSTS guidelines with
and ultrasound endoscopy followed by mediastinoscopy. No differences were found between the mediastinoscopic and endoscopic techniques in terms of sensitivity and negative predictive values.
However, the combination of ultrasound endoscopy for mediastinal staging significantly improved the accuracy of N2/3 staging compared to mediastinoscopy alone. Another result suggests that ultrasound endoscopy for mediastinal lymph node staging can significantly reduce the need for mediastinoscopy.
On the other hand, the results showed that patients who underwent negative ultrasound endoscopy were 13-15% more likely to present with pathologically confirmed N2 involvement postoperatively. This percentage is not low enough in our opinion. Therefore, we still recommend patients to undergo other staging methods at the same time, in order to avoid negative ultrasound endoscopy with postoperative pathologically confirmed N2 involvement.
However, there are prospective studies published by experienced centers that suggest that after performing an ultrasound endoscopic puncture of at least 3 stations of mediastinal lymph nodes, assuming a mediastinoscopy will not further improve the sensitivity of staging. EBUS-TBNA and EUS puncture is a safe technique with complications of <1%.
With the rapid development of these techniques, there are frequent reports of serious complications such as pneumothorax requiring closed drainage, bronchial cyst infection, emphysema, pulmonary/mediastinal abscess, and mediastinal hematoma, among others. However, only one death associated with the EBUS-TBNA technique has been reported so far.
3. Surgical staging techniques
3.1 Transcervical mediastinoscopy
Transcervical mediastinoscopy is performed through an incision in the anterior tracheal fascia of the superior sternal fossa, as proposed by Carlens in 1959. It allows the surgeon to obtain a complete picture of the lymph nodes in the ipsilateral and even contralateral mediastinum. Transcervical mediastinoscopy should be performed under general anesthesia and is also a very safe procedure in the outpatient setting. Many years later, it remains the gold standard for mediastinal staging techniques in patients with operable lung cancer.
Since 1995, the development of television-assisted techniques has also been applied to mediastinoscopy, called television-assisted mediastinoscopy (VAM).VAM can significantly improve the visualization and teaching techniques of mediastinoscopic techniques, as both trainer and trainee can view the operating screen on a monitor.
There are several retrospective studies comparing the advantages and disadvantages of conventional mediastinoscopy with VAM. Although some scholars found that VAM allowed for a greater number as well as more stations of lymph nodes to be obtained, there was no difference in negative predictive values between the two. In some studies, a significant decrease in the incidence of VAM complications has been found.
Just recently, a good body of evidence confirmed the difference in safety and accuracy between VAM and conventional mediastinoscopy. The authors analyzed 108 articles published between 1989 and 2011, of which 5,156 were conventional mediastinoscopy and 959 were VAM. Both techniques were safe, with no fatalities in this time frame and low complications. Although VAM allows for a higher number of lymph node stations, the negative predictive values are identical.
Although the TV mediastinoscopy technique does not always allow obtaining thorough and clinically satisfactory results, it still has many advantages over conventional mediastinoscopy: it has larger and clearer images, it allows sharing the procedure to trainers and other scholars, as well as it can record valuable information for future teaching and research, and it can also improve the quality of teaching while ensuring the safety and accuracy of the examination.
Not only that, it allows direct lymph node clearance, not just biopsy and sampling. This is a very feasible technique for subserosal lymph nodes. After removal of 7 groups of lymph nodes, the esophagus can be very clearly exposed. Therefore the recommendation level of the ESTS working group for VAM is high.
3.2 Television-assisted thoracoscopic technique (VATS)
Although VATS allows access to almost all stations of lymph nodes, it is more invasive than mediastinoscopy (requiring at least two operating holes), and it is limited by thoracic adhesions and does not allow assessment of contralateral mediastinal involvement. In contrast, for groups 5 and 6 lymph nodes, a larger tissue specimen can be obtained with VATS on the left side.
If group 5 or 6 lymph nodes are suspiciously positive on PET, these lymph nodes are difficult to obtain by conventional mediastinoscopy and EBUS is an alternative to VATS for left-sided mediastinal evaluation. In some experienced centers
In some experienced centers, extensive mediastinal evaluation can also be performed through a mediastinoscopic incision, and it is also possible to obtain 5 or 6 groups of lymph nodes with a negative predictive value of 0.89-0.97.
3.3 Television-assisted mediastinoscopic lymph node dissection (VAMLA) Transcervical mediastinoscopic expanded lymph node dissection (TEMLA)
Over the past decade, two more radical invasive staging techniques have gradually entered the picture, they are VAMLA and TEMLA.The aim of these two techniques is to increase the accuracy of staging by providing a more complete resection of the mediastinal lymph nodes and surrounding tissue.VAMLA can be performed using a TV mediastinoscopic incision, whereas TEMLA requires a 5-8 cm incision in the neck and requires suspension of the sternum by hooks.
This technique requires an open approach and is performed with the use of a TV mediastinoscope. With VAMLA, a whole station of lymph nodes such as prevascular, aortic, para-aortic, and paraesophageal lymph nodes can be obtained. In contrast, TEMLA has a negative predictive value of 98.7%. While there is no doubt that these techniques can significantly improve staging accuracy, it is also associated with a high complication and lethality rate.
The postoperative complications of VAMLA and TEMLA have also been well documented and studied, and these techniques are currently only performed in more experienced centers; the main problems with VAMLA are periodic nerve palsy and scar formation, while the problems with TEMLA are mainly lethality and complication rates.
We conclude that the current experience with TEMLA and VAMLA is insufficient and therefore do not recommend their routine use, except in clinical trials. We encourage other centers to share their data on these new staging techniques.
A schematic diagram of preoperative mediastinal staging is shown in the figure. PET or PET-CT is recommended for both mediastinal and distant staging of non-small cell lung cancer.
Summary
Assuming all three criteria are met, surgery can be performed directly: no suspicious lymph node metastases on CT or PET, tumor ≤3 cm and located in the periphery (evidence IIA)
If CT or PET suggests metastatic lymph nodes, histological confirmation is required. EBUS/EUS fine needle aspiration is preferred, after all it is the least invasive and has a high sensitivity to exclude mediastinal involvement (IA evidence). If negative, TV-assisted mediastinoscopy is recommended (IB evidence). The highest accuracy can be obtained with the combination of endoscopic and surgical techniques.
In patients with left lung tumors, lymph nodes in the main pulmonary window should be evaluated using surgical techniques if they are suspicious for involvement (mediastinoscopy, VATS, or mediastinoscopic extension techniques, etc.)
EBUS/EUS/mediastinoscopy is recommended if any of the following points are met: central type lesion, suspicious N1 involvement (evidence IIB). For patients >3 cm (mostly adenocarcinoma patients with increased FDG uptake), the negative predictive value of imaging mediastinal evaluation is <90% and therefore invasive staging is recommended (evidence IIB). Although increased FDG uptake in the primary tumor is a predictor of N2 disease, the ideal threshold value of SUV is currently controversial and therefore invasive assessment is recommended.
In addition to this, SUV measurement is currently not standardized across centers and therefore visualization of the picture is recommended. Among all the above mentioned techniques, local institutions can choose the appropriate one among VAM/EBUS/EUS according to their own experience and medical level, and the basic principle is how to obtain the results of mediastinal assessment using the least invasive method.
If the TV-assisted mediastinoscopy is negative, the patient can proceed to the surgical treatment. And assuming they have undergone EBUS/EUS to confirm that no involved lymph nodes are seen, they can also undergo surgery and information directly, assuming that the number of puncture stations and the number of punctures are up to current standards. Otherwise, a mediastinal evaluation with surgical technique is required first.
Optimal mediastinal lymph node staging is a key step in the integrated multidisciplinary treatment, and we now have these excellent techniques and need to improve the diagnostic accuracy with our experienced hands as well.