Background
In 2009, ESMO decided to supplement the ESMO clinical practice guidelines (GCPs) with “consensus meetings” and to make further recommendations. 2010 saw the first meeting on lung cancer in Lugano and the publication of two consensus meetings.
The second meeting, held in Lugano in May 2013, followed the format of version 1. Four working groups were appointed, each consisting of 8-10 experts from multiple disciplines. A total of 35 experts were involved in the process.
Four specific areas were included as follows: NSCLC pathology and molecular biomarkers; first-, second- and multiline therapy for advanced NSCLC; early-stage NSCLC (stages I-II); and locally advanced NSCLC (stage III).
Prior to the start of the meeting, each working group identifies clinically relevant issues suitable for discussion and provides available literature for reference. The groups made recommendations, which were then presented to the expert panel for full discussion and general consensus. The consensus was published in the 2014 issue of AnnOncology and is compiled below.
Morbidity/Epidemiology
In the late 1970s, four prospective randomized controlled trials (RCTs) of lung cancer screening using a combination of chest X-rays and sputum decidual cytology demonstrated that lung cancer screening was associated with a significant reduction in lung cancer mortality.
However, data from the National Lung Cancer Institute’s National Lung Cancer Screening Trial (NLST), published on November 4, 2010, showed that annual low-dose computed tomography (LDCT) resulted in a significant reduction in lung cancer mortality in specific high-risk populations.
Subsequently, LDCT lung cancer screening was first recommended in specific populations, and in 2011, the International Association for the Study of Lung Cancer (IASLC) conducted a CT screening workshop that brought together experts in the field of lung cancer from around the world to discuss standards and quality control for CT screening.
The workshop suggested that quality control standards must be in place before CT screening is offered, and that organized, individualized screening protocols should be recognized and accepted, and that reporting of results should be standardized.
To reduce the risk of overdiagnosis and overtreatment, standardized protocols for image interpretation and nodal management should be further refined, and positive pathology should be subject to multidisciplinary discussion. To ensure the quality and efficacy of the screening process, clinical, radiological and oncological data should be archived in a database.
The IASLC report also proposes criteria for surgery after screening, recommending that screening be limited to centers where minimally invasive surgery can be performed; that the number of benign diseases resected should be relatively low (<15%); and that anatomic segmental resection with frozen sections of N1 and N2 lymph nodes should be performed for pure hairy glass or partially solid nodes with CT presentation of lung cancer <50 px.
The American Cancer Society (ACS) published its 2013 lung cancer screening guidelines reiterating the 2012 systematic evaluation, emphasizing the requirement that adults screened for lung cancer should be screened at a specialized facility for LDCT screening according to procedures and further seen by a multidisciplinary team skilled in the evaluation, diagnosis, and treatment of abnormal lung lesions. If these conditions are not met, the risk of cancer screening is high.
The systematic evaluation also recommended that physicians should provide a complete description of the benefits, limitations, and risks of LDCT screening, and that smokers should be informed that they remain at risk for lung cancer by continuing to smoke and should be encouraged to quit.
Risk models for lung cancer are currently being developed to determine the ability to screen the optimal lung cancer population and the best screening interval. When the evidence from these models is clear, the NLST will propose target populations and intervals for lung cancer screening.
RECOMMENDATION: LDCT screening reduces lung cancer mortality and can be performed outside of clinical trials, and LDCT screening should be performed at an experienced thoracic oncology center that can provide quality control-specific procedures and that has an established multidisciplinary management program for suspicious nodules.
A smoking cessation program should be offered to individuals when LDCT screening is offered. LDCT screening should not be offered to individuals, but patients requesting screening should be referred to specific procedures, as described above.
Diagnosis
Obtaining a definitive histologic diagnosis prior to treatment is ideal for the treatment of early-stage non-small cell lung cancer. However, inaccessibility of bronchoscopy to these lesions may be the greatest challenge, with population studies demonstrating complication rates of up to 15% reported with transthoracic aspiration biopsy, particularly in middle-aged and elderly individuals, smokers, and those with chronic obstructive pulmonary disease (COPD).
The predicted risk of malignancy can be determined by calculating relevant medical history, smoking history and imaging features of the lung nodules to determine diagnosability or treatability. All of these algorithms have limitations that are partially attributable to population differences.
Recent studies of LDCT lung cancer screening have shown that the combination of nodal volume doubling time (VDT) and/or FDG-PET uptake can reduce the number of benign lesions resected.
However, VDT measurements are not routinely performed outside of clinical trials. differences may exist between CT screening populations and lung cancer populations, and therefore, the use of existing VDT or FDG-PET algorithms alone to establish an early NSCLC diagnosis is not currently supported.
Instead, it is recommended that indeterminate isolated pulmonary nodules (SPNs) should be evaluated by a specialized multidisciplinary oncology team, considering all factors related to patient, epidemiology, and surgery, and applying existing guidelines for the evaluation of pulmonary nodules, such as the Fleischner Society, which has been extended to subsolid nodules.
Specialists on multidisciplinary oncology teams can assess the likelihood of benign disease in their populations by using this calculation that has been validated in that population. However, the malignancy calculus should not be used for the clinical assessment of LDCT screening lung nodules and existing SPN assessment criteria must be applied.
In principle, any radical treatment of lung cancer mostly needs to be based on a histological diagnosis and any reasonable histological diagnosis should be attempted before surgery.
In cases where the nodule resembles a malignant tumor according to current diagnostic algorithms and/or where the preoperative diagnosis is unsuccessful or too dangerous, an experienced multidisciplinary team may recommend surgery according to the principle of minimal trauma.
The location, size and solid component of the nodule need to be considered in the assessment of its benignity or malignancy, as well as the best surgical approach.
Recommendation: pathological diagnosis before recommending surgical intervention. In some patients with clinical stage I/II lesions where a pathologic diagnosis is not available preoperatively, an experienced multidisciplinary team with a high suspicion of malignancy through clinical and imaging evaluation may be effective.
A large proportion of patients with early-stage non-small cell lung cancer cannot be treated surgically due to comorbidities and age. Population studies have shown that patients with early-stage cancer are less likely to receive a pathologic diagnosis of lung cancer than those with advanced non-small cell lung, and the elderly and patients with comorbidities are less likely to receive a pathologic diagnosis of lung cancer.
Obtaining a histologic diagnosis is more challenging in patients who are unsuitable or in a critical state of inoperability than in those who are suitable for surgery.
For operable patients, the American College of Chest Surgeons (ACCP) guidelines suggest that preoperative diagnosis is not recommended when the probability of malignancy exceeds 65%.
Many data support the use of stereotactic ablative radiotherapy (SABR) for inoperable patients without histologic confirmation of the tumor, but requires evaluation by a multidisciplinary team of specialists.
Population data also support the notion that patients undergoing SABR have poorer survival outcomes if a pathologic diagnosis cannot be established indicating that the patient may have extensive comorbidities.
Reassuringly, the final diagnosis of benign disease in this population is only 6% or less than for resected tumors, with or without a preoperative diagnosis of local control of the tumor and disease recurrence similar in SABR patients.
Although, for patients without a pathologic diagnosis “nonsurgical biopsy and/or surgical resection, except where specific contraindications exist”.
The former ACCP guidelines recommended SABR at 65% probability of malignancy, however, this guideline recommends SABR at 85% probability of malignancy, which is consistent with the International Association for the Study of Lung Cancer (IASLC), which states that at CT lung cancer screening centers, the final pathologic diagnosis of benign disease should not exceed 15%.
Recommendation: Routine attempts to obtain a pathologic diagnosis are required prior to performing SABR. When the risk of tissue sampling is too great, there should be at least an 85% probability of malignancy according to accepted criteria.
Staging and risk assessment
Risk is a continuous outcome whose probability is usually expressed as 0% to 100%; clinically, the relative value of defining what patients are ‘highly’ most suitable for the population depends on the individual (usually for unknown reasons or difficult to estimate). The definition of a clinical trial or guideline as a ‘high’ value should be briefly described.
The risk needs to be related to a specific outcome that is meaningful. However, a simple principle is also messy in the respiratory research literature (especially exercise testing), using a combination of several study endpoints (e.g., death, pneumonia, and arrhythmias) that are difficult to interpret.The CALGB9238 trial is a prospective multicenter study to validate the use of primary exercise VO2 measures to predict surgical risk. Patients with peak exercise VO2 <65% of predicted values or <16 ml/kg/min were indeed more likely to have complications and a poorer prognosis (respiratory failure or death).
The authors conclude that their data provide a multicenter validation of the use of exercise VO2 as a preoperative assessment in patients with lung cancer, however, the positive approach may provide safety assurance for surgery in some patients, and 58 patients who did not meet operable criteria underwent surgical resection. They had a mortality rate of 2% and twice the survival rate of those who were not operable.
Therefore, while cardiopulmonary exercise testing can be used to screen at-risk patients, further discussion with patients by a multidisciplinary oncology team is necessary. For example, the risk of in-hospital mortality can be estimated using a validated scoring method such as the Thoracoscore. each risk model should be validated, such as the Goldman Cardiac Risk Index recommended by the American Heart Association/American Heart Association (ACC/AHA), which was recalibrated for use in the lung resection population and validated in subsequent applications.
Recommendation: The validated specific risk model can be used to estimate postoperative mortality and complication rates.
When discussing surgical resection of lung cancer, not only resectability but also functional feasibility, especially cardiopulmonary function, needs to be considered.
To assess cardiac risk, it is recommended to use the revised cardiac risk index (RCRI), which has recently been revised and is called the recalibrated chest RCRI. the index is calculated using 4 weighting factors, and patients are classified into 4 classes as risk increases. The index has recently been externally validated.
The European Respiratory Society (ERS) and the European Society of Thoracic Surgeons (ESTS) collaborative task force established clinical guidelines suitable for patients with radically treated lung cancer (surgery and radiotherapy). If FEV1 or DLCO is <80%, exercise testing and fractionated lung function are recommended to determine the maximum extent of resectability. For sublobar resection (large wedge resection or segmental lung resection), no definitive functional criteria are available. The effect of volume reduction can also be taken into account, especially in patients with heterogeneous emphysema.
Recommendations: A precise assessment of cardiopulmonary function is needed before considering surgical resection to estimate the risk of surgical complications.
For cardiac assessment, a readjusted chest RCRI is recommended. respiratory function assessment of FEV1 and DLCO is required; for either index <80%, exercise testing and fractional lung function are recommended. VO2max can be used in this subset of patients to measure exercise capacity and predict postoperative complications.
Treatment of early stage I and II lung cancer
Although, lobectomy remains the standard of care for early stage T1N0 lung cancer, anatomic segmental resection or large wedge resection is now being reconsidered for small, noninvasive or microinvasive lesions, especially those with gross glassy image (GGO) features.
Two recent reviews and a meta-analysis have shown that carefully selected patients achieved similar survival and recurrence rates to lobectomy using sublobar resection, especially for adenocarcinoma in situ ≤50px. Definitive recommendations can only be formulated based on the results of future large randomized controlled trials.
Some specific subgroups of early adenocarcinoma may not all require systemic lymph node dissection. A recent analysis of the Italian COSMOS screening study suggests that systemic lymph node dissection can be avoided in clinical N0 lung cancer when the maximum standardized uptake value on PET scan is <2,0 and when the pathological nodes are ≤10 mm.
Lung cancer surgery may have a pulmonary decompression effect in patients with non-homogeneous emphysema complicated by lung cancer and lesions located in the lesion portion. A “COPD guideline” suggests that patient selection is better in this situation. A number of surgical treatment options are available, as well as a specific preoperative evaluation algorithm.
Recommendation: It is generally accepted that sublobar resection is acceptable for pure vitreous lesions, carcinoma in situ or microinvasive adenocarcinoma. Lobectomy remains the standard surgical treatment option for solid tumors presenting as ≤50px on CT. Patients with emphysema and limited lung function can be observed with lung reduction by resection of lung cancer and emphysema.
Open surgery, televised thoracoscopic surgery (VATS) and robotic surgery for early stage non-small cell lung cancer
A meta-analysis summarizing 21 controlled studies through 2012, including two randomized controlled studies and 19 non-randomized controlled studies. The results showed no differences in in-hospital pulmonary outcomes or mortality regardless of the form of surgery. The authors highlighted a reduced rate of systemic recurrence (i.e., improved disease-free survival, DFS) in patients who underwent minimally invasive lobectomy.
However, most of the studies were non-randomized controlled studies, and the improvement in DFS may have been caused by case-selection bias.The new 2012 version of the study showed lower in-hospital mortality as well as shorter hospital stays in patients who underwent VATS lobectomy. There were no randomized controlled trials comparing robotic surgery with open or thoracoscopic surgery. Many case series have reported good outcomes for robotic surgery. A case-control study of robotic and thoracoscopic lobectomy reported similar results.
In conclusion, evidence from high-level randomized controlled studies comparing VATS and open thoracic surgery is scarce. There is no high-quality evidence for comparative outcomes between robotic, VATS, or open surgery. To date, the small size of most case-control studies limits external validity (the vast majority are single-center studies). Options for open, thoracoscopic, and robotic surgery for early-stage non-small cell lung cancer
Recommendation: Surgeons may choose the appropriate surgical approach open or VATS based on their experience.
Multiple primary lung cancer
Data on the surgical treatment of multiple primary lung cancers are primarily derived from retrospective analyses. With this in mind, the current evidence supports surgery as the treatment of choice for patients with multiple primary lung cancer, either ipsilateral or bilateral.
Recent studies on multiple nodes undergoing surgical resection have shown that the majority of patients with two concurrent tumors and no lymph node metastases have a proven 5-year survival rate of >50%.
However, the 5-year survival rate decreases with the degree of lymph node involvement in multifocal lung cancer, and total resection of the lesion is usually not advocated due to poor postoperative prognosis in N2 patients. In addition to lymph node involvement, a recent pooled analysis of data based on 467 patients with multifocal lung cancer who underwent multilobar lung resection showed poor prognostic factors: advanced age, male gender, and unilateral distribution of the tumor.
Patients with bilateral lung cancer appear to have a better prognosis because this group of patients is more likely to be those with true multiple primary lung cancer and to benefit mostly from surgery due to non-metastatic disease. These prognostic factors should be considered when determining surgical resectability. Although, it seems reasonable for the main tumor to undergo lobectomy to remove small nodules for sublobar resection, however, there is no consensus on the best type of surgical treatment for multiple primary lung cancers.
If surgery is not feasible, other approaches such as local ablation (SABR) and/or systemic therapy should be considered, however, scientific data are lacking. Therefore, especially for the latter, all treatment decisions should be made after discussion by the multidisciplinary oncology team.
Recommendations: If possible, complete resection is recommended. If complete resection is not possible, the choice of more alternative therapies such as local ablation (e.g. SABR) and/or systemic treatment should be discussed by the multidisciplinary oncology team.
Other factors guiding adjuvant therapy
Indications should be further discussed by the multidisciplinary oncology team, taking into account host factors such as age, comorbidities, physical status (PS), and time since surgery and pathology reports. According to data reported in clinical trials, age per se is not a selection factor.
Patients with severe comorbidities were excluded from clinical trials, and data from the Ontario Cancer Registry suggest that adjuvant chemotherapy produces adverse effects mainly in patients with severe comorbidities (Charlson score 3+) who are still suitable for chemotherapy.
Available data show that patients with PS0-1 benefit from adjuvant chemotherapy, while PS2 is rare. The precise interval for starting adjuvant chemotherapy in clinical trials is not yet clear.
Patients included prior to randomization in the partial grouping trial (IALT) were restricted to 60 days after resection. The Ontario, Canada, registry had a more rigorous time setting, with no differences between the 2 groups (0-10 versus 11-16 weeks).
Postoperative radiotherapy should be considered if R1 is resected (positive margins, chest wall) [III, B]. Even though these patients were not included in randomized controlled trials, adjuvant chemistry is recommended for resected R1 patients regardless of lymph node status. If chemotherapy and radiotherapy are required, radiotherapy should be placed after chemotherapy.
The addition of chemotherapy may be considered after radiotherapy in stage II-N1 patients. Although, this has not been reasonably evaluated in clinical studies, it may be similar to the possible benefit in stage II-N1 patients with surgical resection.
Recommendation: The decision should be made by the multidisciplinary oncology team after evaluation of preexisting comorbidities, physical status and postoperative time. Current knowledge suggests that molecular analysis should not guide adjuvant treatment selection, e.g., ERCC1 or gene mutation testing.
Local ablation (SABR)
The results of SABR have been extensively documented in the literature. However, remedial surgery after SABR has only been reported sporadically. A recent Japanese series reported that local recurrence or new primary lung cancer occurring after SABR was common (∼40% after 3 years), with approximately half of the patients receiving remedial therapy. The very limited current experience seems to support the feasibility of surgery after SABR. However, 25% of patients who initially declined surgery in one series of studies underwent SABR. in some cases, surgery after SABR is associated with SABR-related complications.
Acute complications of SABR, such as skin irritation, fatigue, or cough, usually temporary, occur in 5%-40% of patients. Late complications are less common, such as radiation pneumonia, chest wall pain or rib fractures, hemoptysis or bronchial stenosis or necrosis. Therefore, chest wall morbidity and pulmonary toxicity after SABR are to be incorporated into the decision-making process for secondary surgery with preexisting comorbidities. Histologic diagnosis of lung cancer is key to subsequent treatment, regardless of whether emergency or elective surgery is performed after SABR.
Recommendations: Patients with complications after SABR may be offered remedial surgery if feasible. patients who progress after SABR may be offered remedial surgery using the same indications as the initial surgery, although surgery may be more difficult due to the higher surgical risk.
Follow-up
The incidence of primary LDCT in high-risk lung cancer patients is as low as 1% person/year, but this approach has been shown to reduce lung cancer mortality. A large proportion (20%-40%) of patients with non-small cell lung cancer who undergo radical resection and have pathologic stage IA-IIB present with local or distant recurrence. These patients had a continuous risk ratio with a disease recurrence rate of 6%-7% patient/year in the first 4 years, decreasing to 2% patient/year thereafter. In addition, the risk ratio for the development of a second primary cancer in these patients increased steadily by 1%-3% patient/year in the first three years and did not decrease over time.
A dynamic event study of 1506 patients with resected non-small cell lung cancer showed a significant peak in recurrence around 9 months postoperatively and ending in the second and fourth years.
Based on these results, the surveillance strategy can be recommended for patients with stage I-II non-small cell lung cancer who have undergone radical resection, although there has not been a well-designed randomized controlled trial confirming the impact of this strategy on survival outcomes.
The 2013 ESMOCPGs recommend follow-up every 3-6 months for the first 2-3 years after surgery and less frequently thereafter (annually), with history and physical examination, chest fluoroscopy and CT being reasonable follow-up methods. Based on the above data, monitoring every 6 months for the first 2-3 years with contrast-enhanced spiral CT at 12 and 24 months, followed by annual follow-up including chest CT to detect a second primary tumor, is recommended. Although PET-CT has better sensitivity for detecting recurrent lesions in asymptomatic patients, it is not recommended as there is no proven survival benefit compared to chest spiral CT scans alone.The use of PET-CT when CT detection reveals a suspected lung cancer lesion can be helpful for diagnosis.
Recommendations: Every 6 months for the first 2-3 years with a history, physical examination, and preferably contrast-enhanced spiral CT of the chest recommended at 12 and 24 months, and annually thereafter with a history, physical examination, and CT of the chest to detect a second primary tumor. For follow-up, PET-CT is not recommended.
In Europe, close post-treatment follow-up is important, considering that SABR is still a relatively new technology. The frequency of follow-up and imaging surveillance should be based on the experience of each center and also needs to take into account the patient’s own wishes and suitability for remedial treatment. The incidence of radiological changes in the lungs detected by early and late chest CT ranges from 54%-79% to 80%-100%, respectively. Late changes resemble disease recurrence, but only a small percentage of patients have local recurrence detected by biopsy or further imaging. the time point for clinical application of FDG-PET-CT detection after SABR is not yet clear.
PET-CT is generally performed in patients with suspected recurrence after SABR at the time of chest spiral CT. however, indefinite recurrence signs PET for moderately hyper-metabolically active lesions persisting for 2 years after treatment should be interpreted with caution. The relationship between optimal SUVmax threshold and high risk of recurrence remains to be confirmed, and evidence is very limited due to the low local recurrence rate in the available literature. A growing body of evidence from retrospective studies suggests that SUVmax values greater than 5 at 6 months or more after SABR are associated with a high risk of local recurrence. However, due to the presence of false positives for PET, patients suitable for remedial therapy should first undergo biopsy.
RECOMMENDATION: In centers where SABR has recently been performed, patient follow-up is recommended per ESMOCPG2013 plus CT every 6 months for 3 consecutive years, with baseline treatment-related acute/late side effects and local control rates compared to the existing literature.
For individual patients, follow-up by ESMOCPG2013 and for patients suitable for remedial therapy (e.g., surgery, local ablative therapy) CT examinations every 6 months for 3 consecutive years are recommended.
The frequency of follow-up for patients not suitable for remedial therapy can be developed on an individual basis. the time point for clinical application of FDG-PET monitoring after SABR has not been defined and is therefore not recommended.
Selective use of FDG-PET is recommended when recurrence after SABR is suspected on spiral CT of the chest. due to the large number of false positive FDG-PET results, patients suitable for remedial therapy should be biopsied whenever possible.