Abstract

Surgery remains an essential element of the multimodality radical treatment of patients with early-stage nonsmall cell lung cancer. In addition, thoracic surgery is one of the key specialties involved in the lung cancer tumour board. The importance of the surgeon in the setting of a multidisciplinary panel is ever-increasing in light of the crucial concept of resectability, which is at the base of patient selection for neoadjuvant/adjuvant treatments within trials and in real-world practice. This review covers some of the topics which are relevant in the daily practice of a thoracic oncological surgeon and should also be known by the nonsurgical members of the tumour board. It covers the following topics: the pre-operative selection of the surgical candidate in terms of fitness in light of the ever-improving nonsurgical treatment alternatives unfit patients may benefit from; the definition of resectability, which is so important to include patients into trials and to select the most appropriate radical treatment; the impact of surgical access and surgical extension with the evolving role of minimally invasive surgery, sublobar resections and parenchymal-sparing sleeve resections to avoid pneumonectomy.

Tweetable abstract

Surgery represents a key element in the multidisciplinary management of lung cancer. Surgical aspects such as operability, resectability, surgical access and extent of resection are important and should be learned by all nonsurgical members of the MDT. https://bit.ly/41WDYmx

Introduction

Surgery remains an essential element of the multimodality radical treatment of patients with early-stage nonsmall cell lung cancer (NSCLC). As a consequence, the surgeon is one of the essential members in the tumour board. Their importance in the setting of a multidisciplinary panel is ever-increasing in light of the crucial concept of resectability, which is at the base of patient selection for neoadjuvant/adjuvant treatments.

In this review we will cover some of the topics which are relevant in the daily practice of a thoracic surgeon and should also be known by the nonsurgical members of the tumour board: the pre-operative fitness evaluation of the surgical candidate in light of the ever-improving nonsurgical treatment alternatives unfit patients may benefit from; the definition of resectability which is critical to select the most appropriate radical treatment; and the impact of the surgical access and surgical extension with the evolving role of minimally invasive surgery (MIS), sublobar resections and parenchymal-sparing sleeve resections to avoid pneumonectomy.

MethodsSearch strategy

The search strategy was as follows: original talks addressing the subject from the Collaborative Course of the European Society of Thoracic Surgeons (ESTS) and the European Respiratory Society (ERS) on Thoracic Oncology held in Zurich (February 2023) and their relevant references were included. Additional searches for each of the subtopics included in this narrative review were conducted to retrieve relevant publications which were not discussed during the presentation at the aforementioned meeting. Peer-reviewed studies in English from January 1990 to June 2023 from PubMed and MEDLINE databases were evaluated with regard to the immediate affinity of the published papers to the original talks. Letters to the Editor, congress abstracts and case reports were excluded.

It is not our aim to provide an extensive and systematic review of the topics presented here. Instead, this review aims to further our understanding of the most recent expert opinion and related evidence in pre-operative selection of the surgical candidate, definition of resectability, impact of the surgical access on outcome, and parenchymal-sparing resections for early-stage and more locally advanced-stage lung cancer.

Pre-operative risk assessment of the surgical candidate

Early and localised NSCLC is a heterogenous potentially curable group with various tumour stages, biology and varying prognosis. Treatment options range from segmentectomy in early lung cancer to radiotherapy plus systemic therapy in a multimodality setting in more advanced disease. To improve oncological outcome, patients are often treated with various treatment modalities. Their toxicities and functional impairments can overlap and be increased.

Comorbidities and risks in lung cancer patients

Major issues can be bronchopulmonary and cardiovascular as well as liver, renal and neurological diseases, diabetes and alterations of the immune system [1–3]. Scarce evidence regarding consequences can be found for many comorbidities; however, there are specific recommendations for COPD and cardiovascular diseases.

For interstitial lung diseases (ILDs), literature mostly from Asia describes increased surgical complications, mostly pulmonary adverse events, and early deaths [4–6]. Also, for stereotactic and conventional radiotherapy, increased pulmonary complications are reported [7–9]. Acute ILD exacerbations have a high risk of death (25–100%).

Patients with specific comorbidities (HIV, dialysis, severe immunosuppression, organ transplant, severe autoimmune disease and pulmonary hypertension) should be treated in centres able to provide radical treatment as well as experience in supporting these comorbidities.

Unwanted effects of treatment

For surgery, general anaesthesia, mechanical ventilation and partial collapse of the lung are needed. This impacts primarily the respiratory system, reducing lung volumes, particularly functional residual capacity, and causing post-operative complications [10]. Resection causes loss of lung volume, impaired breathing mechanics and pain, which can increase complications.

Radiotherapy can induce radiation pneumonitis and eventually lung volume loss, and cardiac damage [9, 11–13].

Some systemic treatments such as chemotherapies, e.g. induction chemotherapy, can impair lung function [14–16]. Immunotherapy can cause pneumonitis [17]. Intercurrent complications such as infections due to immunosuppression or allergies can further compromise the patient.

Pulmonary function and prognosis

Pre-operatively, forced expiratory volume in 1 s (FEV1) is usually measured. A correlation with pulmonary complications and mortality is accepted. Pre-operative diffusing capacity of the lung for carbon monoxide (DLCO) is separately predictive of morbidity and mortality [18, 19]. It is also predictive of toxicity for chemoradiotherapy [20]. The calculated, estimated post-operative values for FEV1, DLCO and peak oxygen uptake (peak V′O2) can give additional information [21, 22].

Exercise tests evaluate overall cardiopulmonary function, and partly also coordination, metabolism and muscle strength. In particular, peak V′O2 and the slope of minute ventilation to carbon dioxide production (V′E/V′CO2) during ergo-spirometry are predictive for post-operative complications and mortality [23–27].

Prognostic factors and possible harm by treatments

The natural course of lung cancer is influenced by performance status [28]. It is used as an important decision factor for treatment. There are also several risk scores published, which are not generally used.

The concepts of frailty and nutrition in oncology are also relevant to lung cancer. The prevalence of frailty in lung cancer is 45%, and frailty increases mortality (hazard ratio 3.01) [29] and therapeutic toxicity (odds ratio 2.60) [30]. For patients aged >60 years, frailty phenotype assessment can be considered [31].

Malnutrition is prevalent in lung cancer patients and impacts post-operative complications and 1-year mortality [32]. For patients with underweight or unintentional loss of 10% of their body weight, the application of the Nutritional Risk Index or Nutritional Risk Score can be considered.

Guidelines for assessing the risk of surgical candidates

Some guidelines have been published for the evaluation for curative therapy. Regarding surgery, especially the ERS/ESTS and American College of Chest Physicians guidelines have to be mentioned [33, 34]. They focus on cardiological and pulmonary risk factors and were published several years ago. Meanwhile, further improvements of peri-operative management (especially enhanced recovery after surgery (ERAS)) and thoracic techniques (video-assisted thoracoscopic surgery (VATS) and robot-assisted thoracic surgery (RATS)) have occurred [35–37]. Many treatments are now multimodal, including immunotherapy and advanced radiation techniques.

An Expert Panel of the American Association for Thoracic Surgery Clinical Practice Standards Committee recently generated a list of important risk factors in the determination of high risk for lobectomy. The most important factors for high risk were the need for supplemental oxygen because of severe underlying lung disease, low DLCO, the presence of frailty, and the overall assessment of daily activity and functional status [38]. A Task Force of the ERS/ESTS is currently updating its guidelines.

Risk reduction by preventive measures and optimisation of general conditions

For surgery, several options for risk minimisation exist: smoking cessation, optimising current treatment of comorbidities, pre- and post-operative rehabilitation, and optimisation around surgery (ERAS, minimal invasive thoracic surgery (MITS) and sublobar resection).

Those who quit smoking ∼4 weeks before surgery have a reduced risk of post-surgical complications [39–41].

In terms of pre- and post-operative rehabilitation, patients who are taught breathing exercises pre-operatively or who strengthen their respiratory muscles before surgery with a loaded resistive breathing device can halve their risk of developing post-operative pulmonary complications after major surgery [10]. Pre-operative exercise training improves physical fitness and reduces the risk of post-operative pulmonary complications while minimising hospital resources utilisation [42]. Exercise training improves exercise capacity and quadriceps muscle force, and probably quality of life, and decreases dyspnoea following lung resection for NSCLC [43, 44].

How to choose the right treatment

Multidisciplinary team discussions must consider the patient's wishes and the individual situation as well as the available options. The best treatment regarding the tumour may not be the best option for the patient. For example, in real life, 408 out of 686 (59%) stage III lung cancer patients were considered as ineligible for concurrent chemoradiation [45]. Experienced centres and interdisciplinary specialist discussion are important, at least in patients with increased risk. For example, patients with resection for early-stage NSCLC at facilities with higher use of stereotactic body radiotherapy (SBRT) showed lower rates of post-operative mortality [46].

Definition of resectability in locally advanced disease

Complete resection is the main aim of a cancer operation. It has been clearly defined for lung cancer surgery; however, in locally advanced disease, there are peculiarities when mediastinal lymph nodes or anatomical structures are involved. Complete resection is defined as microscopically tumour-free resection margins, systematic nodal dissection or lobe-specific systematic nodal dissection, no extracapsular nodal extension of the tumour and no metastasis in the highest mediastinal node. If any of the aforementioned criteria are not achieved the resection should be regarded as incomplete. Other criteria that define incomplete resection are microscopically positive pleural or pericardial effusions [47]. Uncertain resection is defined as the presence of carcinoma in situ at the bronchial margin or positive pleural lavage cytology while the resection margins are free and no residual tumour is left [47]. In an oncologically acceptable operation, ESTS guidelines recommend removal of, at least, three hilar and interlobar nodes and three mediastinal nodes from three stations in which the subcarinal is always included [48]. Complete resection and lymph node negativity are proven prognostic factors of long-term survival after resection for locally advanced lung cancer. Thus, pre-resection invasive staging of the hilar and mediastinal lymph nodes is recommended in these patients. Despite radiologically negative findings, 13–15% lymph node positivity is found in post-resection specimens [49].

T-based resectability

The 8th edition of the TNM (tumour, node, metastasis) staging system of lung cancer describes locally advanced disease as a T3 or T4 tumour with intrapulmonary, hilar or mediastinal lymph node involvement [50]. A T3 tumour is >5 cm but ≤7 cm in greatest dimension or associated with separate tumour nodule(s) in the same lobe as the primary tumour or directly invades the chest wall (including the parietal pleura and superior sulcus tumours), phrenic nerve and parietal pericardium. A T4 tumour is >7 cm in greatest dimension or associated with separate tumour nodule(s) in a different ipsilateral lobe than that of the primary tumour or invades the diaphragm, mediastinum, heart, great vessels, trachea, recurrent laryngeal nerve, oesophagus, vertebral body and carina [50].

Bronchus

There are a few studies suggesting a 1.5–2 cm bronchial resection margin [51, 52]; however, these suggestions have not been tested in prospective randomised studies and would not be considered strong evidence. Tumour positivity can be observed in the epithelium, stroma and adventitia of the bronchus [53]. Extramucosal spread has been described as a poor prognostic factor [54]. Safe margins are more frequently achieved after neoadjuvant treatment [53].

Chest wall

Chest wall invasion is categorised according to the depth of invasion, namely parietal pleura, bony structures and soft tissues. Depth of invasion is correlated with prognosis, meaning deeper invasion leads to poorer prognosis [55]. Similarly, a 2 cm clean resection margin is recommended for lung cancers with chest wall invasion [56]. This not always achievable, especially if the invasion is in the posterior paravertebral area or thoracic inlet.

Others

Complete en bloc resection with clear margins is associated with favourable prognosis in T4 tumours with invasion to the carina, vertebra, vena cava, atrium and aorta [57, 58]. However, these are very challenging cases that require expertise and most of those cases are performed in experienced centres [55, 58–61]. The results of those series are reported in table 1. Neoadjuvant treatment was shown to improve complete resection rates and is thus highly recommended in such cases [58]. There is no recommendation for margin distance; however, a wider margin would lead to a lower local recurrence rate.

TABLE 1

Important series in the literature evaluating resection of locally advanced tumours

N-based resectability

There has been no advantage of complete mediastinal lymph node dissection in early-stage lung cancer if the sampling of ipsilateral mediastinal lymph nodes was negative for metastasis [62]. It is also evident that with increasing T stage, the likelihood of mediastinal lymph node metastasis is higher. Thus, a complete mediastinal lymph node dissection is recommended for accurate staging and complete resection in locally advanced lung cancer [48].

N1 disease

Multiple N1 or N1 station (peripheral/hilar) metastases were associated with poor prognosis compared with single N1 or N1 station metastasis [63]. Direct invasion to N1 lymph nodes had better survival compared with hilar lymph node metastasis [64]. There is no proven benefit in terms of locoregional recurrence and long-term survival by performing a pneumonectomy if the metastatic lymph nodes can be completely resected with lesser resections and clean margins [63, 64].

N2 disease

Complete resection for N status is defined as nonmetastatic lymph nodes in the highest lymph node station of the ipsilateral side [48]. This can be interpreted as 2R in the right side and 4L in the left side if a pre-operative mediastinoscopy or endobronchial ultrasound is not performed to evaluate 2L. Surgical resection in lung cancer with N2 disease has not been associated with improved survival [65].

However, single-station N2 disease with upper lobe tumours or mediastinal lymph node clearance after neoadjuvant treatment was associated with significantly longer survival rates [66, 67]. Multiple N2 or bulky N2 disease is associated with poor survival and a decision should be made in a multidisciplinary board meeting for these borderline resectable cases [68, 69].

Considerations after neoadjuvant treatment

Neoadjuvant treatment in various forms is intended to achieve responses in the primary tumour. A much higher rate (almost 30–50%) of major or complete pathological response is observed in locally advanced lung cancers following neoadjuvant chemoimmunotherapy compared with chemotherapy alone [70]. Handling of the tumour specimen to determine margins is based on a series of recommendations by a multidisciplinary group [71]. “Tumour bed” is the preferred term to describe the major components, namely viable tumour, necrosis or stroma which can include inflammation or fibrosis. It is still not clear how the resection status should be defined if there is a complete response (necrosis or inflammation/fibrosis) of the tumour at the surgical margin.

Impact of surgical access (open and MIS (VATS and RATS)) on outcome

Surgical resection of lung cancer can be accomplished using various thoracic accesses. The most commonly used are thoracotomy (also known as the open approach) and MIS access, either by VATS or RATS. Over the years, numerous variations of these approaches have been developed, utilising different numbers of ports, various surgical instruments and devices, CO2 insufflation, nonintubated anaesthesia, etc.

Overview of the evidence

Multiple benefits of the VATS approach have been reported [72]. MIS was found to reduce acute and chronic pain, and thus require a shorter length of hospitalisation [73]. Thoracoscopic lobectomy was associated with less morbidity and mortality, fewer complications, increased quality of life, and a more rapid return to function than lobectomy by thoracotomy [74, 75]. In older patients, more independence after discharge was reported after VATS lobectomy [76]. Thoracoscopic lobectomy was also found to improve the ability of patients to complete post-operative chemotherapy regimens [77]. Overall costs of VATS lobectomy were similar to or lower than open lobectomy [78], while RATS was more expensive and led to longer operating times [79].

Due to their nonrandomised retrospective observational nature, these studies carry a moderate to high bias prevalence that reduces the generalisability of their findings. A recently published meta-analysis of 19 retrospective trials tried to address this problem [80]. Both MIS approaches (VATS and RATS) were found superior regarding near-term mortality in early- and advanced-stage lung cancer and elderly patients. However, benefits were not demonstrated for high-risk patients. Overall survival was similar between open and MIS approaches in all subgroup comparisons. When comparing RATS with VATS, disease-free survival was improved when the robotic approach was used, without any significant difference in overall survival. No statistically significant difference was found regarding overall survival or disease-free survival when comparing RATS with open procedures [80].

Recently, the results of two large randomised controlled trials (RCTs) studying the impact of VATS access were also published [81, 82]. Each trial randomised more than 500 patients.

Long et al. [81] reported that VATS had an advantage in terms of surgery time and bleeding, with similar short-term complications and outcomes compared with thoracotomy. There were no statistically significant differences in chest tube duration, length of hospital stay or complications between the two groups.

Lim et al. [82] reported that VATS led to a better recovery of physical function at 5 weeks, with no difference in physical function at 6 and 12 months, compared with thoracotomy, and no difference in oncological outcomes such as disease-free survival and overall survival. Patients experienced less in-hospital pain with less analgesic consumption after VATS lobectomy. The benefit was sustained on all pain-related quality-of-life subscales (including overall, chest and incision pain) documented for up to 1 year. There was no difference in serious adverse events in the hospital, but safety issues, such as prolonged air leakage and vascular injury, were more common in the VATS group.

A meta-analysis of seven RCTs found that VATS and thoracotomy did not differ in terms of operating time or early mortality [83]. However, hospital stay was slightly shorter in VATS compared with thoracotomy. The two approaches also did not differ in terms of bleeding, prolonged air leakage, acute respiratory failure or arrhythmia. However, the certainty level was low due to a limited number of complications and significant measurement variations.

Interpretation

There is now a worldwide consensus and a recommendation that MIS techniques should be preferred for the surgical treatment of lung cancer in patients with no anatomical or surgical contraindications as long as there is no compromise of the standard oncological and dissection principles of thoracic surgery [84].

Since current evidence indicates that the benefit of MIS is moderate and short lived, a more prudent recommendation for the use of MIS in lung cancer surgery may be more appropriate, particularly when utilising MIS in patients with more significant technical challenges and risks of surgical complications, such as in advanced lung cancer, post-neoadjuvant chemoimmunotherapy [71] and complex pulmonary resections with bronchovascular reconstructions [85].

There is a significant learning curve associated with VATS lobectomies [86]. When surgeons insufficiently proficient in MIS are suggested to obsessively utilise this technique to avoid the nonrecommended open approach, they may unintentionally compromise clinical outcomes due to prolonged surgery time, increased bleeding [87] or even catastrophic complications [88].

Considering the even longer learning curve associated with anatomical segmental resections [89], thoracotomy may be still considered in some complex situations for this procedure.

Over the past decades, the open technique has advanced along with MIS. New devices such as endo-staplers, advanced energy devices, novel surgical instruments, intra-operative magnification and three-dimensional reconstruction have increased the efficiency of open lung cancer surgery and reduced the size of the thoracotomy, procedure time and post-operative complications. The utilisation of ERAS has significantly reduced post-operative morbidity and hospital stay in open lung cancer surgery. At the same time, the benefit of ERAS in VATS was relatively minor [90]. In a multivariate comparison of predictors of prolonged hospital stay, the benefit of ERAS in lung cancer surgery was roughly 2-fold compared with the benefit of VATS [91].

Sublobar versus lobar resection for early-stage NSCLC

An anatomical lobectomy has long been established as the “gold standard” for curative resection of early-stage lung cancer [92–94]. However, the potential of sublobar resection to better preserve lung function compared with lobectomy continued to prove alluring to surgeons [95, 96] and clinical experience accumulated over the last couple of decades has now identified niches where such limited resection may be an alternative, or even possibly preferable, to lobectomy.

Compromised patients

Because the volume of lung removed is reduced, sublobar resection logically should most benefit “compromised” patients. A 2016 multicentre study from Japan on patients with clinical stage I lung cancer who were deemed unfit for lobectomy found that sublobar resection could yield a 3-year overall survival rate of 79.0% and a 3-year recurrence-free survival rate of 75.9% [97]. Similar results were found in another Japanese study on patients deemed unfit for lobectomy [98]. Patients with fewer than two “high risk” factors and undergoing a sublobar resection had a 3-year recurrence-free survival rate of 95%. A 2014 systematic review further found that sublobar resection for patients with clinical stage IA lung cancer yielded mortality rates of 0–1.4%, 5-year overall survival rates of 40–79% and local recurrence rates of 5.2–5.5% [99]. While these figures are generally inferior to those expected for lobectomy in noncompromised patients, they are nonetheless acceptable for patients who cannot receive lobectomy. Moreover, a number of studies have suggested that sublobar resection offered similar morbidity rates as SBRT but possibly better mid- to long-term survival [100, 101]. Such clinical evidence has now led to sublobar resection (especially segmentectomy) being recognised in major guidelines today as appropriate in compromised patients with poor pulmonary reserve or other major comorbidities that contraindicate lobectomy [102, 103].

Intentional sublobar resection

Surveillance Epidemiology End Results (SEER) database results demonstrate significantly worse survival for small (≤2 cm) NSCLC patients after intentional sublobar resection than after lobectomy during 1987–1997 [102]. However, over time, the difference progressively reduced. By 2005–2008, the difference between sublobar resection and lobectomy was no longer significant [102]. It is probable that this phenomenon may be the result of two areas of progress: refinement of sublobar surgery and better patient selection for sublobar resection.

Surgical considerations

Many observational studies over the past two decades have consistently demonstrated that anatomical segmentectomy offers superior survival to nonanatomical wedge resections when treating stage IA lung cancer [103, 104]. This may be related to the increased likelihood of segmentectomy achieving wide resection margins compared with wedge resection. A study from 2007 found that local recurrence after sublobar resection was significantly more likely if the resection margin was <1 cm [105] and a study from 2013 also found that a wedge resection was significantly more likely to result in a resection margin <1 cm than a segmentectomy [106]. Nevertheless, the recent CALGB 140503 randomised trial showed equal 5-year survival between sublobar resection and lobectomy for cT1aN0 lung cancer even though the sublobar arm included 58.8% patients receiving wedge resections [107]. This has generated considerable discussion over whether wedge resection may eventually prove equivalent to segmentectomy.

The other area where surgical technique may influence the results of sublobar resection is the strategy for lymph node dissection. Several observational studies have consistently demonstrated that with sublobar resection for clinical stage I lung cancer, survival is significantly worse in patients where no lymph node dissection or sampling had been performed [108, 109]. In risk factor analyses for sublobar resection of lung cancer, lymph node dissection and upstaging have also been shown to be correlated with oncological outcomes [110, 111]. An elegant study from 2011 showed that sublobar resection (mostly wedge resections) without lymph node sampling resulted in significantly worse overall and recurrence-free survival than lobectomy for small (≤2 cm) NSCLC [112]. However, when sublobar resection was performed with lymph node sampling, that difference compared with lobectomy was erased. There is currently broad consensus that nodal dissection should be performed for sublobar resection for lung cancer. However, much greater controversy exists over what should be done if N1 nodal metastasis is found during a sublobar resection. One school of thought argues that conversion to a lobectomy is mandatory [113]. However, the opposing school of thought posits that lobectomy is unnecessary, pointing to a number of studies that have repeatedly shown that lobectomy offers no better survival than segmentectomy should unexpected N1 nodal metastasis be found [114–116].

Patient selection considerations

The appearance of the target cancer on computed tomography (CT) scanning as a ground-glass opacity (GGO) or lesion with a consolidation-to-tumour ratio (CTR) <50% generally indicates suitability for resection by sublobar resection. In risk factor analyses for sublobar resection of lung cancer from Japan, such CT appearance has been shown to be correlated with oncological outcomes [117, 118]. A systematic review of 14 cases series from Japan looking at resection of GGOs or lesions with CTR ≤50% (studies included 52–100% of all resections by wedge resection or segmentectomy) found that both 5-year overall and 5-year disease-free survival rates ranged from 93% to 100%, with a local recurrence rate of 0% [119]. It is probable that such GGO or GGO-dominant lesions may represent pre-malignant or less aggressive forms of adenocarcinoma [120]. The Japanese JCOG0804 multi-institutional confirmatory phase III trial evaluated the use of sublobar resection for peripheral GGO-dominant lung cancer in 314 patients (82% wedge resections) and confirmed a 5-year recurrence-free survival rate of 99.7% (95% CI 97.7–100.0%), with no local relapse [121]. Current guidelines recognise pre-malignant histology and/or CTR <50% as criteria for choosing to electively perform sublobar resection [93].

The other key criterion in selecting for sublobar resection is a tumour size ≤2 cm. A meta-analysis of 22 observational studies comparing segmentectomy versus lobectomy for stage I lung cancer in 2013 noted that the hazard ratios of overall and cancer-specific survival indicated significant benefits of lobectomy for stage I, stage IA and stage IA with tumours 2–3 cm in size: 1.20 (95% CI 1.04–1.38; p=0.011), 1.24 (95% CI 1.08–1.42; p=0.002) and 1.41 (95% CI 1.14–1.71; p=0.001), respectively [122]. However, if only tumours ≤2 cm were considered, segmentectomy provided an effect equivalent to that of lobectomy: hazard ratio 1.05 (95% CI 0.89–1.24; p=0.550). The most important clinical evidence to date of course comes from the two major randomised trials comparing sublobar resection versus lobectomy. The Japanese JCOG0802 trial looked at 1106 patients with peripheral tumours ≤2 cm and with CTR >50% (solid-dominant) and found that segmentectomy was noninferior to lobectomy in terms of 5-year overall survival [36]. The North American CALGB 140503 trial similarly looked at peripheral tumours ≤2 cm and found that sublobar resection (58.8% wedge resections) was again noninferior to lobectomy in both 5-year disease-free and 5-year overall survival rates [107].

Other indications

The rise in interest in sublobar resection coincides with the advent of the era of low-dose CT screening for lung cancer, which has the potential to reduce lung cancer mortality more than any known treatment [123–126]. However, current guidelines on how to manage screening-detected lesions tend to recommend conservative “wait-and-see” approaches for almost all such lesions [93]. This caution arises from a concern of over-diagnosis (some lesions may not be cancerous), which in turn stems from the perception that surgical intervention (traditionally lobectomy) is relatively traumatic [127]. Modern sublobar resection performed via MIS access approaches promises to significantly reduce the morbidity of intervention [128]. There is therefore a school of thought that now suggests a more proactive strategy for intervening and even resecting screening-detected lesions using sublobar resection, lowering thresholds for considering surgery [129, 130].

Another increasingly recognised indication for sublobar resection is multifocal lung cancer [131–133]. In such situations, sublobar resection becomes important to achieve multiple curative resections without excessive sacrificing of the patient's lung function [133, 134].

Parenchymal-sparing sleeve resections or pneumonectomy

Sleeve resections of the bronchial tree and/or the pulmonary arterial vasculature are established parenchymal-sparing resection techniques for centrally located NSCLC with airway or vascular invasion as well as in tumours directly originating from the airway. Whenever feasible, parenchyma-sparing resection techniques are preferable over pneumonectomy due to superior peri-operative and long-term outcomes as well as quality of life. Sleeve resections are feasible from the level of the trachea to the segmental level with various reconstruction techniques available.

Pre-operative work-up should always include bronchoscopic assessment and consideration of risk factors for anastomotic healing beyond standard imaging and functional tests [135].

No prospective randomised trial comparing survival after sleeve lobectomy versus pneumonectomy has been conducted and probably never will be. Thus, approximation of benefit can best be obtained by large, matched cohorts. A recent propensity score-matched analysis retrospectively compared 665 patients undergoing sleeve lobectomy with 665 patients undergoing pneumonectomy (9.7% of the sleeve resections were combined bronchovascular procedures) [136]. Complication rates (3.61% versus 8.72%) as well as 30- and 90-day mortality (0.6% and 1.5% versus 0.9% and 3.91%) were lower in the sleeve group; 5-year overall survival (61% versus 44.7%) as well as 5-year disease-free survival (56.6% versus 46.2%) were superior in the sleeve cohort. Another recent study from the same institution compared exclusively 139 bronchovascular sleeve resections with pneumonectomy and found similar results [137].

Functional and oncological outcomes after sleeve lobectomy are comparable to standard lobectomy [138, 139], even though a higher peri-operative mortality has been described by a single retrospective analysis which found 5-year survival rates to be comparable again [140].

The ESTS database, one of the largest international thoracic surgical databases, was analysed for patients undergoing sleeve lobectomy or bilobectomy [141]. 1652 patients were found eligible for analysis. 161 (9.7%) underwent a VATS procedure. Neoadjuvant treatment was given in 24.1% of patients without increased morbidity or mortality. The overall mortality was low (2.2%) and complication rates were within the expected range. Bronchial anastomotic complications occurred in 1.8% and in the majority of cases (79.8%) were managed surgically. The minimally invasive VATS approach was associated with a decreased post-operative complication rate (odds ratio 0.64, 95% CI 0.42–0.98) in multivariable analysis and a shorter length of stay (median (interquartile range) 5 (3–8) versus 8 (7–12) days). The conversion rate, however, was relatively high at 21.1%. Other recent publications confirm the reduced number of complications after MIS [142] and the noninferiority in survival rate of a VATS approach compared with open surgery [143].

While sleeve lobectomy is a safe procedure, several technical issues can be debated. While some centres advocated routine coverage of the bronchial anastomosis, in retrospective data of several institutions it has been shown that neither the number of anastomotic complications nor the peri-operative mortality or long-term outcome is influenced by wrapping the anastomosis. This is also valid for the subgroup of patients receiving neoadjuvant chemotherapy or chemoradiotherapy [144]. This also corresponds to the observations in lung transplantation irrespective of the actual suture technique used [145, 146]. A single running suture technique is feasible for sleeve resections as well as interrupted stitches on the cartilaginous portion.

Most of the available single-centre analyses report peri-operative complication rates and mortality in sleeve resections after neoadjuvant chemotherapy or chemoradiotherapy comparable to primary surgery [147–149]. Adding neoadjuvant immunotherapy neither increases peri-operative morbidity nor mortality [150–152] and does not seem to impair anastomotic healing. Only limited data are available on the best timing of sleeve resection after neoadjuvant therapy [153], thus no firm conclusion can be drawn at this stage.

Particularly for low-grade malignancies and tumours directly originating from the central airways, several advanced parenchymal-sparing resection and reconstruction techniques including neo-carinal formation have been described; however, these are mainly limited to individual case reports or small case series [154–158].

Points for clinical practice or future research

Pre-operative selection of the surgical candidate. The flow of the decision includes: 1) assessment of respiratory and cardiovascular risk as well as general patient condition and comorbidities; 2) evaluation of physical activity/exercise capacity and options for improvement; 3) specific therapy options according to tumour stage, biology and patient situation; 4) assumed oncological outcome (multidisciplinary discussion); 5) overall evaluation of risks versus benefit (multidisciplinary discussion) and shared decision making with the patient; and 6) re-evaluation after each therapy step.

Importance of the surgical access. The ideal surgical access should be individualised to a particular lung cancer resection. It should provide an optimal balance of safety, speed, cost-efficiency, and short- and long-term oncological outcomes. Selection should be based on the patient's characteristics, the surgeon's skills, and the whole surgical and hospital environment. The relatively moderate short-term benefits of MIS over thoracotomy should always be weighed against its potential downsides. Sometimes, this can only be decided during the procedure. Therefore, any conversion from MIS to thoracotomy should never be seen as a failure, only performed in case of a vascular injury with severe bleeding. Instead, conversion should be made as a conscious decision to use thoracotomy to provide the patient with the best outcome of lung cancer resection.

Define resectability in locally advanced lung cancer. The following criteria might be used in clinical practice: 1) ensure a 2 cm margin on the chest wall and safe distance from the tumour on the airway/bronchus considering peribronchial extension of the tumour; 2) consider tumour-free margins that can be achieved safely for T4 tumours (no distance recommendations); 3) there is no need for pneumonectomy if interlobar, hilar lymph nodes are completely removed; 4) perform removal of lymph nodes based on lymph node staging guideline recommendations; and 5) these cases should always be discussed at multidisciplinary meetings (in addition, a higher rate of neoadjuvant or peri-operative treatment should be pursued to achieve higher complete resection rates).

Lobar versus sublobar resections for early-stage lung cancers. Accumulation of clinical data internationally has now identified several areas where sublobar resection is justified for the curative treatment of early-stage lung cancer. In compromised patients who cannot receive lobectomy, sublobar resection offers acceptable oncological outcomes which may be superior to nonsurgical modalities. In elective situations, sublobar resection appears appropriate for GGO and GGO-dominant lesions ≤2 cm in size, preferably with segmentectomy and lymph node dissection performed. The latest randomised trial data suggest indications may be extended to even solid and solid-dominant cancers ≤2 cm in size. The management of screening-detected lesions and multifocal lung cancer are niches where sublobar resection appears to be gaining increasing importance.

Parenchymal-sparing resections for locally advanced lung cancer. Bronchial and bronchovascular sleeve resections are established routine procedures in the treatment of NSCLC as well as in other tumours originating from the central airways. Parenchyma-sparing techniques should be used whenever feasible to achieve a complete resection. Sleeve resections can be safely performed after neoadjuvant therapy including regimens containing immune checkpoint inhibitors as well as by MIS approaches in selected patients.

Footnotes

Provenance: Commissioned article, peer reviewed.

Conflicts of interest: The authors report the following conflicts of interest in the past 36 months. C. Aigner participated in an advisory board with and received personal consulting fees from AstraZeneca, BMS, MSD, Roche and Ewimed; and received institutional grants from BMS, Ewimed, PharmaCept, Medtronic and D-A-CH medical group. H. Batirel participated in an advisory board with and received personal consulting fees from AstraZeneca, Johnson & Johnson and Medtronic; and is the Secretary General of the ESTS. R.M. Huber participated in an advisory board with AstraZeneca, Novocure, BMS, Bayer, Roche, Boehringer Ingelheim, Beigene, Sanofi, Janssen and Merck; and is a co-chair of the ERS/ESTS Task Force on “Fitness for radical therapy in lung cancer” and IASLC Deputy Chair of the Screening and Early Detection Committee. D.R. Jones has no conflicts to declare. A.D.L. Sihoe received speaker honoraria from AstraZeneca, Medela, Medtronic and Roche; and is a member of the Board of Directors of the STS, ASCVTS and Asia Thoracocospic Surgery Education Platform. T. Štupnik received speaker honoraria from Johnson & Johnson, Roche and MSD; and participated in an advisory board with AstraZeneca and Johnson & Johnson. A. Brunelli participated in an advisory board with and received personal consulting fees from AstraZeneca, BMS, MSD, Roche, Johnson & Johnson and Medtronic; and is a co-chair of the ERS/ESTS Task Force on “Fitness for radical therapy in lung cancer” and member of the Board of Directors of the ESTS and STS.

Received October 9, 2023.Accepted January 11, 2024.Copyright ©The authors 2024http://creativecommons.org/licenses/by-nc/4.0/

This version is distributed under the terms of the Creative Commons Attribution Non-Commercial Licence 4.0. For commercial reproduction rights and permissions contact permissionsersnet.org