Surgery, RT, and systemic therapy are the 3 modalities most commonly used to treat patients with NSCLC. They can be used either alone or in combination depending on the disease status. In the following sections, the clinical trials are described that have led to the recommended treatments.
In general, for patients with stage I or II disease, surgery provides the best chance for cure.
Thoracic surgical oncology consultation should be part of the evaluation of any patient being considered for curative local therapy. The overall plan of treatment and the necessary imaging studies should be determined before any nonemergency treatment is initiated. It is essential to determine whether patients can tolerate surgery or whether they are medically inoperable; some patients deemed inoperable may be able to tolerate minimally invasive surgery and/or sublobar resection.
Although frailty is an increasingly recognized predictor of surgical and other treatment morbidity, a preferred frailty assessment system has not been established.
The Principles of Surgical Therapy are described in the NSCLC algorithm and are summarized here (see the NCCN Guidelines for NSCLC). Determination of resectability, surgical staging, and pulmonary resection should be performed by board-certified thoracic surgeons who should participate in multidisciplinary clinics and/or tumor boards for patients with lung cancer. Surgery may be appropriate for select patients with uncommon types of lung cancer (eg, superior sulcus, chest wall involvement) (see the NCCN Guidelines for NSCLC).
Patients with pathologic stage II or greater disease can be referred to a medical oncologist for evaluation. For resected stage IIIA, consider referral to a radiation oncologist. Treatment delays, because of poor coordination among specialists, should be avoided. The surgical procedure used depends on the extent of disease and on the cardiopulmonary reserve of the patient. Lung-sparing anatomic resection (sleeve lobectomy) is preferred over pneumonectomy, if anatomically appropriate and if margin-negative resection can be achieved; lobectomy or pneumonectomy should be done if physiologically feasible.
Sublobular resection, either segmentectomy (preferred) or wedge resection, is appropriate in select patients; the parenchymal resection margins are defined in the NSCLC algorithm (see Principles of Surgical Therapy in the NCCN Guidelines for NSCLC).
Resection (including wedge resection) is preferred over ablation.
Wide wedge resection may improve outcomes.
Patients with medically inoperable disease may be candidates for SABR, also known as stereotactic body RT (SBRT).
If SABR is considered for patients at high risk, a multidisciplinary evaluation is recommended (see Stereotactic Ablative Radiotherapy in this Discussion).
Lymph Node Dissection 淋巴结清扫
A randomized trial (ACOSOG Z0030) compared systematic mediastinal lymph node sampling versus complete lymphadenectomy during pulmonary resection in patients with either N0 (no demonstrable metastasis to regional lymph nodes) or N1 (metastasis to lymph nodes in the ipsilateral peribronchial and/or hilar region, including direct extension) NSCLC disease. In patients with early-stage disease who had negative nodes by systematic lymph node dissection, complete mediastinal lymph node dissection did not improve survival.
Thus, systematic lymph node sampling is appropriate during pulmonary resection; one or more nodes should be sampled from all mediastinal stations. For right-sided cancers, an adequate mediastinal lymphadenectomy should include stations 2R, 4R, 7, 8, and 9. For left-sided cancers, stations 4L, 5, 6, 7, 8, and 9 should be sampled.
Patients should have N1 and N2 node resection and mapping (American Thoracic Society map) with a minimum of 3 N2 stations sampled or a complete lymph node dissection.
The lymph node map from the IASLC may be useful.
Formal ipsilateral mediastinal lymph node dissection is indicated for patients undergoing resection for stage IIIA (N2) disease. For patients undergoing sublobular resection, the appropriate N1 and N2 lymph node stations should be sampled unless not technically feasible because sampling would substantially increase the surgical risk. Sublobular resection, either segmentectomy (preferred) or wedge resection, is appropriate in select patients (see Principles of Surgical Therapy in the NCCN Guidelines for NSCLC): 1) those who are not eligible for lobectomy; and 2) those with a peripheral nodule 2 cm or less with very low-risk features. Segmentectomy (preferred) or wedge resection should achieve parenchymal resection margins that are: 1) 2 cm or more; or 2) the size of the nodule or more.
The role of surgery in patients with pathologically documented stage IIIA (N2) disease is described in the NSCLC algorithm (see Principles of Surgical Therapy in the NCCN Guidelines for NSCLC) and summarized here. Before treatment, it is essential to carefully evaluate for N2 disease using radiologic and invasive staging (ie, EBUS-guided procedures, mediastinoscopy, thorascopic procedures) and to discuss whether surgery is appropriate in a multidisciplinary team, which should include a board-certified thoracic surgeon.
Neoadjuvant (preoperative) therapy is recommended for select patients. The optimal timing of RT in trimodality therapy (preoperative with chemotherapy or postoperative) is not established and controversial.
In patients with N2 disease, 50% of the NCCN Member Institutions use preoperative chemoradiotherapy whereas 50% use preoperative chemotherapy.
There is no evidence that adding RT to induction regimens improves outcomes for patients with stage IIIA (N2) disease when compared with using chemotherapy alone.
Clinicians also agree that resection is not appropriate for patients with multiple pathologically proven malignant lymph nodes greater than 3 cm; definitive chemoradiotherapy is recommended for these patients. The NCCN Panel believes that surgery may be appropriate for select patients with N2 disease, especially those whose disease responds to induction chemotherapy (see Principles of Surgical Therapy in the NCCN Guidelines for NSCLC).
It is controversial whether pneumonectomy after preoperative chemoradiotherapy is appropriate.
Patients with resectable N2 disease should not be excluded from surgery, because some of them may have long-term survival or may be cured.
Thorascopic Lobectomy 胸腔镜叶切除术
Video-assisted thoracic surgery (VATS), which is also known as thorascopic lobectomy, is a minimally invasive surgical treatment that is currently being investigated in all aspects of lung cancer (see Principles of Surgical Therapy in the NCCN Guidelines for NSCLC).
Published studies suggest that thorascopic lobectomy has several advantages over thoracotomy.
Acute and chronic pain associated with thorascopic lobectomy is minimal; thus, this procedure requires a shorter length of hospitalization.
Thorascopic lobectomy is also associated with low postoperative morbidity and mortality, minimal risk of intraoperative bleeding, or minimal locoregional recurrence.
Thoracoscopic lobectomy is associated with less morbidity, fewer complications, and more rapid return to function than lobectomy by thoracotomy.
In patients with stage I NSCLC who had thorascopic lobectomy with lymph node dissection, the 5-year survival rate, long-term survival, and local recurrence were comparable to those achieved by routine open lung resection.
Thorascopic lobectomy has also been shown to improve discharge independence in older populations and patients at high risk.
Data show that thorascopic lobectomy improves the ability of patients to complete postoperative chemotherapy regimens.
Based on its favorable effects on postoperative recovery and morbidity, thorascopic lobectomy (including robotic-assisted approaches) is recommended in the NSCLC algorithm as an acceptable approach for patients who are surgically resectable (and have no anatomic or surgical contraindications) as long as principles of thoracic surgery are not compromised (see Principles of Surgical Therapy in the NCCN Guidelines for NSCLC).
The Principles of Radiation Therapy in the NSCLC algorithm include the following: 1) general principles for early-stage, locally advanced, and advanced NSCLC; 2) target volumes, prescription doses, and normal tissue dose constraints for early-stage, locally advanced, and advanced NSCLC; and 3) RT simulation, planning, and delivery.
These RT principles are summarized in this section. Whole brain RT and stereotactic radiosurgery (SRS) for brain metastases are also discussed in this section. The abbreviations for RT are defined in the NSCLC algorithm (see Table 1 in Principles of Radiation Therapy in the NCCN Guidelines for NSCLC).
Treatment recommendations should be made by a multidisciplinary team. Because RT has a potential role in all stages of NSCLC, as either definitive or palliative therapy, input from board-certified radiation oncologists who perform lung cancer RT as a prominent part of their practice should be part of the multidisciplinary evaluation or discussion for all patients with NSCLC. Uses of RT for NSCLC include: 1) definitive therapy for locally advanced NSCLC, generally combined with chemotherapy; 2) definitive therapy for early-stage NSCLC in patients with contraindications for surgery; 3) preoperative or postoperative therapy for selected patients treated with surgery; 4) therapy for limited recurrences and metastases; and/or 5) palliative therapy for patients with incurable NSCLC.
The goals of RT are to maximize tumor control and to minimize treatment toxicity. Advanced technologies such as 4D-conformal RT simulation, intensity-modulated RT/volumetric modulated arc therapy (IMRT/VMAT), image-guided RT, motion management strategies, and proton therapy have been shown to reduce toxicity and increase survival in nonrandomized trials.
A secondary analysis of a randomized trial (RTOG 0617) reported that 2-year overall survival, PFS, local failure, and distant metastasis-free survival were not significantly different for IMRT when compared with 3D-conformal RT. IMRT yielded lower rates of severe pneumonitis when compared with 3D-conformal RT (3.5% vs. 7.9%; P = .039).
CT-planned 3D-conformal RT is now considered to be the minimum level. Definitive RT, particularly SABR, is recommended for patients with early-stage NSCLC (ie, stage I-II, N0) who are medically inoperable or those who refuse surgery (see Stereotactic Ablative Radiotherapy in this Discussion).
SABR is also an option for patients at high surgical risk who cannot tolerate a lobectomy (eg, major medical comorbidity or severely limited lung function). Resection is recommended for patients with early-stage NSCLC who are medically fit (see Principles of Surgical Therapy in the NCCN Guidelines for NSCLC).
Definitive chemoradiation is recommended for patients with stage II to III disease who are not appropriate surgical candidates.
Involved-field RT (also known as involved-field irradiation or IFI) is an option for treating nodal disease in patients with locally advanced NSCLC; IFI may offer advantages over elective nodal irradiation (ENI).
For patients with advanced lung cancer (ie, stage IV) with extensive metastases, systemic therapy is recommended; palliative RT can be used for symptom relief and potentially for prophylaxis at primary or distant sites.
Shorter courses of palliative RT are preferred for patients with symptomatic chest disease who have poor PS and/or shorter life expectancy (eg, 17 Gy in 8.5 Gy fractions) (see Table 4 in the Principles of Radiation Therapy in the NCCN Guidelines for NSCLC). Higher dose and longer course thoracic RT (eg, ≥30 Gy in 10 fractions) are associated with modestly improved survival and symptoms, especially in patients with good PS.
The indications for using preoperative or postoperative chemoradiation or RT alone are described in the NSCLC algorithm (see Principles of Radiation Therapy in the NCCN Guidelines for NSCLC). In patients with clinical stage I or II NSCLC who are upstaged to N2+ after surgery, postoperative chemotherapy can be administered followed by postoperative RT (also known as PORT) depending on the margin status (see the NCCN Guidelines for NSCLC).
For clinical stage III NSCLC, definitive concurrent chemoradiation is recommended (category 1). The optimal management of patients with potentially operable stage IIIA NSCLC is controversial and is discussed in detail in the algorithm (see Principles of Surgical Therapy in the NCCN Guidelines for NSCLC).
For patients undergoing preoperative therapy before surgical resection of stage IIIA NSCLC, some oncologists prefer chemotherapy alone rather than chemoradiotherapy for the preoperative treatment; RT should generally be given postoperatively if not given preoperatively. The NCCN Panel recommends a preoperative RT dose of 45 to 54 Gy.
NCCN Member Institutions are evenly split in their use of preoperative chemotherapy versus preoperative chemoradiation in patients with stage IIIA N2 NSCLC.
在ⅢA N2 NSCLC患者中，使用术前化疗与术前放化疗的NCCN成员机构各占一半。
Similarly, some consider the need for pneumonectomy to be a contraindication to a combined modality surgical approach given the excess mortality observed in clinical trials, but NCCN Member Institutions are split on this practice as well. Surgery is associated with potentially greater risk of complications, particularly stump breakdown and bronchopleural fistula, in a field that has had high-dose RT (eg, 60 Gy). Thus, surgeons are often wary of resection in areas that have previously received RT doses of more than 45 to 50 Gy, especially patients who have received definitive doses of concurrent chemoradiation (ie, ≥60 Gy) preoperatively. Soft tissue flap coverage and reduced intraoperative fluid administration and ventilator pressures can reduce the risk of these complications.
When giving preoperative RT to less than definitive doses (eg, 45 Gy), one should be prepared up front to continue to a full definitive dose of RT without interruption if the patient does not proceed to surgery for some reason. For these reasons, when considering trimodality therapy, the treatment plan—including assessment for resectability and the type of resection—should be decided before initiation of any therapy.
Target Volumes, Prescription Doses, and Normal Tissue Dose Constraints
The dose recommendations for preoperative, postoperative, definitive, and palliative RT are described in the Principles of Radiation Therapy in the NSCLC algorithm (see Table 4 in the NCCN Guidelines for NSCLC).
After surgery, lung tolerance to RT is much less than for patients with intact lungs. Although the dose volume constraints for conventionally fractionated RT for normal lungs are a useful guide (see Table 5 in Principles of Radiation Therapy in the NCCN Guidelines for NSCLC), more conservative constraints should be used for postoperative RT. The NCCN Panel noted that the doses and constraints provided in the tables are not specific prescriptive recommendations; they are useful reference doses that have been commonly used or are from previous clinical trials. For definitive RT, the commonly prescribed dose is 60 to 70 Gy in 2 Gy fractions over 6 to 7 weeks.
The use of higher RT doses is discussed in the NSCLC algorithm (see Principles of Radiation Therapy in the NCCN Guidelines for NSCLC).
Doses more than 74 Gy are not currently recommended for routine use.
Results from a phase 3 randomized trial (RTOG 0617) suggest that high-dose radiation using 74 Gy with concurrent chemotherapy does not improve survival, and might be harmful, when compared with a dose of 60 Gy.
Although the RT dose to the heart was decreased in the RTOG 0617 trial, survival was decreased; thus, more stringent constraints may be appropriate. Reports 50, 62, and 83 from the International Commission on Radiation Units and Measurements provide a formalism for defining RT target volumes based on grossly visible disease, potential microscopic extension, and margins for target motion and daily positioning uncertainty (see Figure 1 in Principles of Radiation Therapy in the NCCN Guidelines for NSCLC); the ACR Practice Parameters and Technical Standards are also a helpful reference.
It is essential to evaluate the dose volume histogram (DVH) of critical structures and to limit the doses to the organs at risk (such as spinal cord, lungs, heart, esophagus, and brachial plexus) to minimize normal tissue toxicity (see Table 5 in Principles of Radiation Therapy).
These constraints are mainly empirical and have for the most part not been validated rigorously.
The QUANTEC review provides the most comprehensive estimates from clinical data of dose-response relationships for normal tissue complications.
As previously mentioned, for patients receiving postoperative RT, stricter DVH parameters should be considered for the lungs.
Radiation Simulation, Planning, and Delivery
Treatment planning should be based on CT scans obtained in the treatment position. Intravenous contrast CT scans are recommended for better target delineation whenever possible, especially in patients with central tumors or nodal involvement. FDG PET/CT can significantly improve target delineation accuracy, especially when there is atelectasis or contraindications to intravenous CT contrast.
In the NSCLC algorithm, recommendations are provided for patients receiving chemoradiation (including those with compromised lung or cardiac function), photon beams, or IMRT (see Radiation Therapy Simulation, Planning, and Delivery in the Principles of Radiation Therapy in the NCCN Guidelines for NSCLC).
Respiratory motion should be managed. The report from the AAPM Task Group 76 is a useful reference for implementing a broad range of motion management strategies as described in the NSCLC algorithm (see Radiation Therapy Simulation, Planning, and Delivery in the NCCN Guidelines for NSCLC).
With conventionally fractionated RT, 3-year survival is only about 20% to 35% in these patients, with local failure rates of about 40% to 60%.
In prospective clinical trials, local control and overall survival appear to be considerably increased with SABR, generally more than 85%, and about 60% at 3 years (median survival, 4 years), respectively, in patients who are medically inoperable.
Substantially higher survival has been observed in patients with potentially operable disease who are treated with SABR; survival is comparable in population-based comparisons to surgical outcomes, but locoregional recurrences are more frequent.
It has not been shown that use of SABR for medically operable patients provides long-term outcomes equivalent to surgery. Late recurrences have been reported more than 5 years after SABR, highlighting the need for careful surveillance.
If possible, biopsy should confirm NSCLC before use of SABR.
SABR is recommended in the NSCLC algorithm for patients with stage I and II (T1–3,N0,M0) NSCLC who are medically inoperable; SABR is a reasonable alternative to surgery for patients with potentially operable disease who are high risk, elderly, or refuse surgery after appropriate consultation (see the NCCN Guidelines for NSCLC).
A combined analysis of 2 randomized trials (that did not complete accrual) assessed SABR compared with lobectomy in operable patients.
The analysis does not alter the fact that surgical resection is recommended and typically used for operable patients, but it helps to confirm the indication of SABR for patients with contraindications for surgery or those who refuse surgery. SABR can also be used for patients with limited lung metastases or limited metastases to other body sites.
After SABR, assessment of recurrences by imaging can be challenging because of benign inflammatory/fibrotic changes that can remain FDG-PET avid for 2 or more years after treatment, emphasizing the importance of follow-up by a team with experience interpreting such post-treatment effects.
SABR fractionation regimens and a limited subset of historically used maximum dose constraints are provided in the NSCLC algorithm (see Tables 2 and 3 in the Principles of Radiation Therapy in the NCCN Guidelines for NSCLC).
These dose constraints are point-of-reference doses and are not intended to be prescriptive; they are used commonly or have been used in clinical trials.
Although none of these dose constraints has been validated as a maximally tolerated dose, outcomes of clinical trials to date suggest that they are safe constraints.
The bronchial tree, esophagus, and brachial plexus are critical structures for SABR. For centrally located tumors—those within 2 cm in all directions of any mediastinal critical structure, including the bronchial tree, esophagus, heart, brachial plexus, major vessels, spinal cord, phrenic nerve, and recurrent laryngeal nerve—regimens of 54 to 60 Gy in 3 fractions are not safe and should be avoided; 4 to 10 fraction SABR regimens appear to be effective and safe (see Principles of Radiation Therapy in the NCCN Guidelines for NSCLC).
Preliminary results (RTOG 0813) suggest that 5-fraction regimens are safe.
SRS or SABR for limited oligometastases to the brain or other body sites, respectively, may be useful for patients with good PS and thoracic disease that can be treated with definitive therapy (see Stage IV, M1b: Limited Sites in the NCCN Guidelines for NSCLC).
Local therapy combined with targeted therapy is a category 2A recommendation for patients with ALK or ROS1 rearrangements or sensitizing EGFR mutations.
Decisions about whether to recommend SABR should be based on multidisciplinary discussion. Hypofractionated or dose-intensified conventional 3D-conformal RT is an option if an established SABR program is not available.
Nonrandomized clinical data indicate that local tumor control with SABR is higher than with interventional radiology ablation techniques. Interventional radiology ablation may be appropriate for selected patients for whom local control is not necessarily the highest priority.
Many patients with NSCLC have brain metastases (30%–50%), which substantially affect their quality of life.
Options for treatment of limited brain metastases include 1) SRS alone; and 2) surgical resection for selected patients followed by SRS or whole brain RT. Selected patients include those with symptomatic metastases or whose tumor tissue is needed for diagnosis (see the NCCN Guidelines for NSCLC).
Treatment of limited brain metastases in patients with NSCLC differs from that recommended in the NCCN Guidelines for Central Nervous System Cancers, because patients with NSCLC and brain metastases often have long-term survival; therefore, the potential neurocognitive issues that may occur with whole brain RT are a concern.
At 3 months after SRS alone, patients had less cognitive deterioration (40/63 patients [63.5%]) than those receiving SRS plus whole brain RT (44/48 patients [91.7%]; difference, -28.2%; 90% CI, -41.9% to -14.4%; P < .001).
Decisions about whether to recommend SRS alone or brain surgery followed by whole brain RT or SRS for limited brain metastases should be based on multidisciplinary discussion, weighing the potential benefit over the risk for each individual patient.
Treatment should be individualized for patients with recurrent or progressive brain lesions.
For multiple metastases (eg, >3), whole brain RT is recommended; SRS may be preferred for patients who have good PS and low systemic tumor burden (see the NCCN Guidelines for Central Nervous System Cancers, available at www.NCCN.org).
Whole brain RT is associated with measurable declines in neurocognitive function in clinical trials, particularly with increasing dose and advanced age of the patient.
However, control of brain metastases confers improved neurocognitive function.
For limited metastases, randomized trials have found that the addition of whole brain RT to SRS decreases intracranial recurrence but does not improve survival and may increase the risk of cognitive decline.
Thus, SRS or whole brain RT alone is recommended for patients with limited volume metastases.
Some have suggested that resection followed by SRS to the cavity (instead of resection followed by whole brain RT) will decrease the risk of neurocognitive problems.
A study suggests that using IMRT to avoid the hippocampus may help decrease memory impairment after whole brain RT.
A phase 3 randomized trial assessed optimal supportive care (including dexamethasone) with whole brain RT versus optimal supportive care alone in patients with NSCLC and brain metastases who were not eligible for brain surgery or SRS.
As previously mentioned, surgery provides the best chance for cure for patients with stage I or II disease who are medically fit and can tolerate surgery. SABR can be considered for patients with unresectable stage I or II (T1–3, N0) disease or those who refuse surgery if their disease is node negative (see Stereotactic Ablative Radiotherapy in this Discussion and see the NCCN Guidelines for NSCLC). In patients with completely resected NSCLC, adjuvant (postoperative) chemotherapy has been shown to improve survival in patients with early-stage disease.
Some studies suggest that preoperative chemotherapy (also referred to as neoadjuvant chemotherapy or induction chemotherapy) is as effective as and better tolerated than postoperative chemotherapy (see Preoperative Chemotherapy Followed by Surgery: Trial Data in this Discussion).
A randomized trial found no difference in survival with preoperative versus postoperative chemotherapy.
The NCCN Guidelines state that patients with stage II or IIIA (T3, N1) disease may be treated with induction chemotherapy before surgery if they are candidates for therapy after surgery.
Concurrent chemoradiation is more efficacious than sequential chemoradiation for patients with unresectable stage III disease.
For patients with stage IV disease who have a good PS, platinum-based chemotherapy is beneficial.
Data show that early palliative care combined with systemic therapy improved quality of life, mood, and survival in patients with metastatic NSCLC, even if these patients had less aggressive end-of-life care, when compared with those not receiving palliative care alone.
Patients should receive treatment for debilitating symptoms.
A study also suggests that social support, such as being married, is as effective as systemic therapy.
Preliminary results from a recent study indicate that systematic symptom monitoring during outpatient chemotherapy treatment increases overall survival when compared with usual care.
Surgery is rarely recommended for patients with stage IV disease. However, surgical resection of limited brain metastases may improve survival in selected patients with stage IV disease and is recommended for selected patients in the NCCN Guidelines (see the NCCN Guidelines for NSCLC, available at www.NCCN.org).
Definitive local therapy with surgical resection or RT is recommended for limited metastases located in sites other than the brain if definitive thoracic therapy is feasible (see Stage IVA, M1b: Limited Sites in the NCCN Guidelines for NSCLC).
The trials supporting the recommendations for combined modality therapy are discussed in the following sections.
Surgery Followed by Chemotherapy: Trial Data
In the NSCLC algorithm for resected stage IA disease, postoperative chemotherapy is not recommended based on the trials described in the following paragraphs.
Postoperative chemotherapy may be considered for high-risk, margin-negative, stage IB disease (see the NCCN Guidelines for NSCLC). Recommended chemotherapy regimens for preoperative and postoperative therapy are provided in the NCCN Guidelines.
For the 2018 update (Version 1), the NCCN Panel added 2 preoperative and postoperative therapy regimens for patients with comorbidities or those not able to tolerate cisplatin, including 1) carboplatin/gemcitabine; and 2) carboplatin/pemetrexed (nonsquamous only).
The International Adjuvant Lung Cancer Trial (IALT) reported a statistically significant survival benefit with cisplatin-based postoperative therapy in patients with completely resected stage I, II, or III NSCLC.
The study included 1867 patients with surgically resected lung cancer who were randomly assigned either to cisplatin-based postoperative chemotherapy or to observation, with a median follow-up duration of 56 months. A higher survival rate (45% vs. 40% at 5 years; HR for death, 0.86; 95% CI, 0.76–0.98; P < .03) and disease-free survival rate (39% vs. 34% at 5 years; HR, 0.83; 95% CI, 0.74–0.94; P < .003) were reported for patients assigned to chemotherapy when compared with observation.
IALT data suggest that cisplatin-based postoperative chemotherapy improves survival 5 years after treatment in patients with completely resected NSCLC. However, after 7.5 years of follow-up, there were more deaths in the chemotherapy group and the benefit of chemotherapy decreased over time.
Data show that postoperative chemotherapy prevents recurrences. The NCIC CTG JBR.10 trial and the ANITA trial compared the effectiveness of postoperative vinorelbine/cisplatin versus observation in early-stage NSCLC. In the JBR.10 trial, 482 patients (ECOG PS of 0–1) with completely resected stage IB (T2a, N0) or stage II (T1, N1, or T2, N1) NSCLC were randomly assigned either to vinorelbine/cisplatin or to observation.
Postoperative chemotherapy significantly prolonged overall survival (94 vs. 73 months; HR for death, 0.69; P = .04) and relapse-free survival (not reached vs. 47 months, HR for recurrence, 0.60; P < .001) when compared with observation alone. The 5-year survival rates were 69% and 54%, respectively (P = .03). When compared with observation alone, postoperative chemotherapy is beneficial for patients with stage II disease but not for stage IB disease as shown by updated data from JBR.10 after 9 years of follow-up.
In patients with stage II disease receiving postoperative chemotherapy, median survival is 6.8 versus 3.6 years in those who were only observed. Of note, patients receiving chemotherapy did not have an increased death rate. In the ANITA trial, 840 patients with stage IB (T2a, N0), II, or IIIA NSCLC were randomly assigned either to postoperative vinorelbine/cisplatin or to observation.
Grade 3/4 toxicities were manageable in the chemotherapy group; 7 toxic deaths were reported. After a median follow-up of 76 months, median survival was 66 months in the chemotherapy group and 44 months in the observation group.
Postoperative chemotherapy significantly improved (8.6%) the 5-year overall survival in patients with completely resected stage II and IIIA disease, although no benefit was observed in stage I. Some clinicians consider vinorelbine/cisplatin to be the preferred regimen for completely resected early-stage NSCLC based on the number of trials and the amount of use; however, most clinicians in the United States prefer to use regimens with less toxicity.
A meta-analysis of 4,584 patients (LACE) found that postoperative cisplatin-based chemotherapy increased survival over 5 years (absolute benefit of 5.4%); there was no difference among the chemotherapy regimens (vinorelbine, etoposide, and others).
A subgroup analysis found that cisplatin/vinorelbine also increased survival.
The benefit was greater in patients with stage II and III disease and with good PS. Postoperative chemotherapy benefited elderly patients up to 80 years of age.
The CALGB 9633 trial assessed paclitaxel/carboplatin in patients with stage IB (T2a, N0, M0) lung cancer.
In this trial, 344 patients were randomly assigned either to paclitaxel/carboplatin or to observation (within 4–8 weeks of resection) with a median follow-up duration of 74 months. Postoperative chemotherapy was well tolerated with no chemotherapy-related toxic deaths. Overall survival at 6 years was not significantly different (although a subset analysis showed a benefit for tumors 4 cm or more), although 3-year survival was significant (80% vs. 73%, P = .02).
Thus, the carboplatin/paclitaxel regimen is only recommended for early-stage disease if patients cannot tolerate cisplatin (see Chemotherapy Regimens for Neoadjuvant and Adjuvant Therapy in the NCCN Guidelines for NSCLC).
It is important to note that the CALGB trial was underpowered for patients with stage 1B disease.
Preoperative Chemotherapy Followed by Surgery:
Data from clinical trials in patients with resected NSCLCs indicate that delivery of chemotherapy is an important problem. In the postoperative setting, significant comorbidities and incomplete recovery after surgery often make it difficult for patients to tolerate systemic therapy. This problem was demonstrated in the NATCH phase 3 trial (which compared surgery alone to preoperative or postoperative chemotherapy with paclitaxel/carboplatin), because 90% of the preoperative cohort completed 3 cycles of chemotherapy but only 61% of the postoperative cohort completed chemotherapy; however, survival was equivalent among all 3 arms.
A randomized trial found no difference in 3-year overall survival (67.4% vs. 67.7%) with preoperative versus postoperative chemotherapy in patients with early-stage NSCLC; response rate and quality of life were similar in both arms.
Postoperative chemotherapy (with or without RT or reresection) is recommended and typically used for early-stage disease in the NCCN Guidelines.
Several trials suggest that preoperative therapy is beneficial in patients with N2 disease.
Other trials suggest that preoperative therapy is beneficial in patients with earlier stage disease.
A follow-up, randomized intergroup trial (SWOG 9900) evaluated preoperative paclitaxel/carboplatin in 354 patients with stage IB to IIIA (but not N2) disease versus surgery alone. The trial closed prematurely because of practice changes and was therefore not appropriately powered. This SWOG trial did show a trend toward improved PFS (33 vs. 20 months) and overall survival (62 vs. 41 months) with preoperative chemotherapy, and no difference in resection rates between the 2 arms.
Scagliotti et al published a phase 3 trial of preoperative cisplatin/gemcitabine versus surgery alone in 270 patients with stage IB to IIIA disease. Although the trial closed early, a significant survival benefit was seen in patients with stages IIB and IIIA disease who received chemotherapy (HR, 0.63).
Song et al published a meta-analysis of all available randomized clinical trials evaluating preoperative chemotherapy in resectable NSCLCs. This meta-analysis evaluated 13 randomized trials; the HR suggests that overall survival in the preoperative chemotherapy arm is similar to the surgery alone arm (HR, 0.84; 95% CI, 0.77–0.92; P = .0001).
The benefit from preoperative chemotherapy is similar to that attained with postoperative chemotherapy.
Chemoradiation: Trial Data
The major controversies in NSCLC relate to the management of patients with stage IIIA disease (see the Role of Surgery in Patients with Stage IIIA (N2) NSCLC in Principles of Surgical Therapy in the NCCN Guidelines for NSCLC). All 3 treatment modalities—surgical resection, chemotherapy, and radiation—may be used when treating stage III disease. The ongoing debate centers on which modalities to use and in what sequence.
For patients with unresectable stage IIIA or stage IIIB disease, combined modality therapy (chemoradiation) is more efficacious than radiation alone.
Concurrent chemoradiation is more efficacious than sequential chemoradiation.
However, concurrent chemoradiation has a higher rate of grade 3 or 4 esophagitis than sequential chemoradiation. Selection of patients should be based not only on the anticipated response to therapy but also on how well the patient is anticipated to tolerate therapy. Frail patients may not be able to tolerate concurrent chemoradiation.
Concurrent chemoradiation regimens that may be used for all histologies for initial treatment include cisplatin/etoposide, cisplatin/vinblastine, and carboplatin/paclitaxel (see Chemotherapy Regimens Used with Radiation Therapy in the NCCN Guidelines for NSCLC).
For nonsquamous NSCLC, additional concurrent chemoradiation regimens may be used including carboplatin/pemetrexed and cisplatin/pemetrexed.
A weekly paclitaxel/carboplatin regimen is another chemoradiation option.
The different options for preoperative, definitive, and postoperative chemotherapy/RT are described in detail in the algorithm.
Recently, the NCCN Panel removed the preferred designation for the cisplatin/etoposide and cisplatin/vinblastine concurrent regimens based on data from a phase 3 randomized trial and a retrospective assessment of the Veterans Administration data.
For the 2018 update (Version 1), the NCCN Panel expanded the list of regimens for sequential chemoradiation to include regimens that are also used for preoperative and postoperative chemotherapy (ie, cisplatin combined with pemetrexed [nonsquamous only], docetaxel, etoposide, gemcitabine, or vinorelbine; carboplatin combined with paclitaxel) and also added 2 new carboplatin regimens for patients with comorbidities or those not able to tolerate cisplatin, including 1) carboplatin/gemcitabine; and 2) carboplatin/pemetrexed (nonsquamous only).
A recent phase 3 randomized trial (PACIFIC) compared consolidation therapy (ie, after chemoradiation) with durvalumab versus placebo in patients with unresectable stage III NSCLC (PS 0–1) who had not progressed after 2 or more cycles of definitive concurrent platinum-based chemoradiation.
Patients received durvalumab after receiving concurrent chemoradiation (1–42 days). Most patients were current or former smokers and did not have EGFR mutations; their PD-L1 status was typically less than 25% or unknown. Durvalumab was effective in patients with both squamous and nonsquamous NSCLC. The PFS was 16.8 months for durvalumab (95% CI, 13.0–18.1) versus 5.6 months for placebo (95% CI, 4.6–7.8) (stratified HR for disease progression or death, 0.52; 95% CI, 0.42–0.65; P < .001).
Grade 3 or 4 adverse events occurred at a similar rate in both groups of patients (durvalumab, 29.9% vs. placebo, 26.1%). Pneumonia was the most common grade 3 or 4 adverse event (durvalumab, 4.4% vs. placebo, 3.8%).
The NCCN Panel recommends durvalumab as consolidation therapy (regardless of PD-L1 status) for patients (PS 0–1) with unresectable stage III NSCLC who have not progressed after 2 or more cycles of definitive concurrent platinum-based chemoradiation based on this trial.
Durvalumab should be discontinued for patients with severe or life-threatening pneumonitis and should be withheld or discontinued for other severe or life-threatening immune-mediated adverse events when indicated (see prescribing information).