Abstract
This article is in our series on thoracic surgery
Introduction
The surgical treatment of non-small-cell lung cancer is evolving new technical advances technologies and algorithms in response to changes in disease presentation which have in turn resulted from demographic changes and advances in early detection. Unfortunately, advances in evidence-based surgical practice will inevitably lag behind pioneering techniques and new protocols. However, one can find new applications for established techniques without compromising surgical standards.
Historical conclusions are no longer applicable
The widely held mantra that the ‘gold standard’ treatment of early lung cancer is lobectomy stems from a randomised trial conducted by the Lung Cancer Study Group in USA over 20 years ago. 1 This trial found that sublobar resection resulted in higher local recurrence and inferior survival when compared to lobectomy. However, the conclusions can be challenged for the following reasons. The fact that the study was conducted over 20 years ago means that clinical staging and preoperative assessment has changed; many patients did not have access to preoperative computed tomography, and positron emission tomography was definitely not available. Thus, there can be no validation of the clinical staging nor was there the rigorous intraoperative staging which is best practice today. There have also been significant advances in surgical techniques with improved technology for the division of lung parenchyma and the use of minimally invasive, video-assisted thoracic surgery. Also, there was too much heterogeneity within the experimental group in the study with a combination of predominantly non-anatomical wedge resection and only a limited proportion of anatomical segmentectomy. As has been found in recent years, results from non-anatomical resections are inferior and more likely to give rise to local recurrence.
The disease and the patients are changing
We are witnessing a transition in the typical patient presenting with operable lung cancer. 2 There has been a shift away from central squamous carcinomas in men towards an increase in smaller peripheral adenocarcinoma in more elderly women, many with co-existent emphysema and compromised lung function. By necessity, lung parenchymal sparing options are becoming the treatments of choice and, fortunately, the new typical tumours lend themselves to minimally invasive techniques.
There is competition for surgery
The advent of alternatives to surgery in the forms of less invasive ablation techniques has been the focus of the attention of surgeons. Rather than simply embarking on a ‘turf war’, surgeons are now prepared to consider less invasive and less extensive surgical alternatives which inevitably will benefit patients. A pooled analysis of two randomised trials (STARS and ROSEL) of stereotactic ablative radiotherapy versus lobectomy for operable stage I non-small-cell lung cancer suggests that in the short term, results are comparable to surgery. 3 However, the standard of surgery in this trial was suboptimal. In many cases, a simple wedge resection was all that was performed and only one in five of the lobectomies was performed by video-assisted thoracic surgery. There was an unusually high survival rate in the stereotactic ablative radiotherapy group (higher than previous series) which may be related to the incomplete histological confirmation rate in the stereotactic ablative radiotherapy group. Furthermore, the study had very limited power to detect differences with short follow-up and few events (31 stereotactic ablative radiotherapy vs. 27 surgery). A larger propensity matched analysis 4 suggests patient benefit from video-assisted thoracic surgery, particularly for larger tumours, in the form of improved cancer-specific survival compared with patients undergoing stereotactic ablative radiotherapy. Unfortunately, the prospective randomised trials of surgery versus stereotactic ablative radiotherapy have consistently failed to recruit.
The surgeon must appreciate segmental anatomy
In order to meet the challenge offered by stereotactic ablative radiotherapy, the modern thoracic surgeon must be technically adept at the demands of anatomical segmental resection, particularly by video-assisted thoracic surgery and therefore must have exemplary anatomical knowledge (Figures 1 and 2). The nomenclature of the anatomy of the bronchopulmonary segments, not clearly established until the report by Brock for the Thoracic Society of Great Britain in 1950,
5
should be common parlance. The numerical labelling of the segments provides a convenient shorthand.
Bronchopulmonary segments of the right lung. Bronchopulmonary segments of the left lung.
While its importance in these early years was in the treatment of infective lung diseases, modern thoracic surgeons must now apply this knowledge to the surgery of lung cancer. There is an argument that because ‘cancer does not respect anatomical boundaries’, thus segmentectomy should never replace lobectomy or that non-anatomical wedge resection is satisfactory. The counter-argument is that in modern lung cancer practice, an increasing proportion of tumours will be diagnosed in stage I, and tumours of this size are unlikely to be large enough to often cross the boundaries of many segments. Nevertheless, it is advisable to respect a macroscopic margin of excision that is at least equal to the diameter of the tumour. Interestingly, it has been estimated from the three-dimensional reconstructions from thin-slice computed tomography scans of virtual tumours that the maximum tumour diameters with a 30 mm parenchymal safety margin ranged from 26.1 mm in right-sided segments 7 + 8 to 59.8 mm in the left apical segments 1–3. 6
The identification of the intersegmental planes can be more of a technical challenge than in bronchiectasis where the target segment is collapsed. Intraoperative reventilation after division of the segmental bronchus is more complicated during video-assisted thoracic surgery as vision is obstructed. Thus more advanced techniques can be employed to identify the correct plane of dissection including intrabronchial instillation of methylene blue 7 or the use of intrabronchial or intravenous Indocyanine green followed by near-infrared thoracoscopy to detect fluorescence.8,9 Preoperative three-dimensional computed pulmonary angiography may facilitate division of the segmental bronchovascular supply and identify any congenital anomalies but problems may still persist in actual cleavage of the parenchyma due to the restricted opening of most endostaplers and the thickness of the lung tissue. The use of electrocautery and other energy devices requires further research.
There is evidence of the efficacy of segmental resection with certain limits
Favourable oncological results can be obtained from anatomical sublobar resections. The prospective randomised trials are yet to report 10 but in a propensity-matched analysis of segmentectomy versus lobectomy for clinical stage I non-small-cell lung cancer, there was found to be no significant difference in freedom from disease or overall survival. 11 In a further non-randomised comparison with lobectomy, sublobar resection had no significant impact on disease-free survival. 12
The importance of resection margin was emphasised by Schuchert et al. 13 who showed the advantages of segmentectomy over lobectomy included reduced operating time and blood loss without increasing mortality or recurrence so long as the margin: tumour diameter was > 1, i.e. a margin > 20 mm for a 2 cm tumour. It appears that histology is also important as in adenocarcinoma, segmentectomy was found to have equivalent survival to lobectomy but in squamous carcinoma neither wedge resection nor segmentectomy were equivalent to lobectomy. 14
When considering sublobar resection, anatomical segmentectomy should be preferred to simple non-anatomical wedge resection as it has been associated with significant improvement in overall and lung cancer-specific survival in a propensity-matched analysis of 3525 patients in the SEER registry. 15
Ideally, as in other lung cancer resections, segmentectomy should be carried out by minimally invasive surgery since compared with thoracotomy it results shorter length of stay, fewer pulmonary complications and yet equivalent lymphadenectomy and similar oncological outcomes. 16 Segmentectomy may be the best indication for robot-assisted thoracic surgery, the intricate dissection facilitated by the improved instrumentation. 17
Where should segmentectomy replace lobectomy?
Segmentectomy is not just for the high-risk individuals with insufficient respiratory reserve to tolerate lobectomy. In a systematic review, segmentectomy offered no significant difference in overall survival in the ‘intentionally selected' group and may be a feasible alternative for selected patients who could tolerate either procedure. 18 There is logic to the argument behind segmentectomy. One would assume that most thoracic surgeons faced with a small right middle lobe tumour would not perform a superior bilobectomy irrespective of the completeness of the horizontal fissure. Thus faced with a similar tumour in the lingula segments of the left upper lobe, particularly with a partially developed lingular fissure, the surgeon would similarly consider a lingulectomy. And if this solitary small tumour arose from the apico-lower segment, then logically a segmentectomy would be considered. Finally, then, if this small tumour actually arose from the anterior basal segment then would not a basal segmentectomy be planned?
This parenchymal sparing approach is especially pertinent in the face of an increasing incidence of multifocal adenocarcinoma which may require further surgical intervention in either lung at a later date. The surgeon would rue his decision to remove the entire lower lobe for a small part-solid lepidic adenocarcinoma when the disease progresses in the contralateral lower lobe and the patient has insufficient pulmonary reserve to undergo further resection.
However, preoperative assessment is paramount as the complexity of a right upper lobe posterior segmentectomy may be misplaced if that lobe is largely emphysematous and its complete removal by lobectomy may be therapeutic. 19
While a rigid adherence to lobar resection should be abandoned, a balance between oncological principles and lung preservation must be agreed. The basic surgical oncological principles include the intention to achieve a R0 resection which has been defined as free resection margins proved microscopically; systematic nodal dissection or lobe-specific systematic nodal dissection; no extracapsular nodal extension of the tumour; and the highest mediastinal node removed must be negative. 20 There is therefore evidence in support of anatomical segmental resection. In small, peripheral tumours (T1a, T1b) which are preferably adenocarcinoma and where a resection margin in excess of tumour diameter can be achieved. One could play devil’s advocate and suggest that segmentectomy should still be considered in N1 disease for where is the benefit in sacrificing functioning lung tissue to control systemic spread? Surely this is best controlled with systemic chemotherapy rather than extended parenchymal resection. The practice of intraoperative frozen section analysis of visible nodes cannot exclude intrapulmonary nodes and is needlessly time-consuming. Segmentectomy is best applied to clinical N0 tumours as identified by preoperative PET-CT and endobronchial ultrasound biopsy if indicated.
In conclusion, one should remember that all chest wall wounds will heal but lung tissue will never grow back. Perhaps, instead of the thoracic surgeon asking ‘which lobe do I need to remove?’, he should preferentially ask ‘how many segments can I safely preserve?’
Keypoints
The surgical treatment of non-small-cell lung cancer is evolving with new technical advances and approaches. The conclusions of the historical gold standard trial are no longer applicable. The disease and the patients are changing. There is competition for surgery. The surgeon must appreciate segmental anatomy. There is evidence of the efficacy of segmental resection with certain limits.