Over the past decade, the treatment of non-small cell lung cancer (NSCLC) has evolved. While early diagnosis and surgical treatment results in optimal patient outcomes, the majority of patients are diagnosed with lung cancer at later, largely incurable stages often requiring multimodality therapy. Over the last decade, we have observed significant improvements in the management of advanced stages primarily due to an increased understanding of the molecular heterogeneity and drivers of lung cancer initiation and progression as well as improvements in radio and surgical therapies. Using high-throughput platforms, investigators have determined that a significant percentage of lung tumors have actionable somatic mutations, each of which represents an opportunity for novel targeted or “personalized” therapy. Investigators have discovered mutations within the epidermal growth factor receptor (EGFR), anaplastic lymphoma kinase (ALK), fibroblast growth factor receptor (FGFR), ROS-1, KRAS, HER2, BRAF, MET, and RET ^{1– 7} . Through clinical trials, distinct roles for targeted therapies are emerging as standard first- or second-line therapies in patients harboring select mutations.

Angiogenesis is one of the hallmarks of cancer, and blockade of vascular endothelin growth factor receptor-2 (VEGFR-2) signaling inhibits the formation, proliferation, and migration of new blood vessels ⁴⁶ . Bevacizumab, a recombinant humanized monoclonal antibody against VEGF, when compared to carboplatin-paclitaxel first-line chemotherapy, resulted in a notable OS improvement in eligible patients with non-squamous NSCLC ⁶⁸ . The anti-VEGFR antibody ramucirumab is a human IgG1 monoclonal antibody that binds to the VEGFR-2 extracellular domain with high affinity, preventing the binding of all VEGF ligands and receptor activation ⁶⁹ . Ramucirumab improved PFS and OS when combined with docetaxel, irrespective of histology, and was approved by the FDA in NSCLC ⁷⁰ . In a multicenter, double-blind, randomized phase III trial, 1,253 patients with squamous and non-squamous NSCLC who progressed during or after a first-line platinum-based chemotherapy regimen were enrolled. Patients were randomized to receive docetaxel plus either ramucirumab or placebo. Median OS was 10.5 months (IQR: 5.1–21.2) among patients randomized to docetaxel plus ramucirumab and 9.1 months (4.2–18) for patients who received placebo plus docetaxel (hazard ratio [HR]: 0.86, 95% CI: 0.75−0.98, P = 0.023). In those given ramucirumab, median PFS was 4.5 months (IQR: 2.3–8.3), while the control group had a PFS of three months (IQR: 1.4–6.9) (HR: 0.76, 95% CI: 0.68–0.86, P <0.0001) ⁷¹ .

Therapies targeting the complex interaction between tumors and the immune system represent the largest recent breakthrough in cancer therapy. Immunotherapy is a particularly attractive option, as it is independent of the cancer cell mutational status and has been proven to be efficacious with a more acceptable side effect profile. Therapies targeting inhibitory checkpoint molecules including cytotoxic T-lymphocyte-associated protein 4 (CTLA4), programmed death 1 (PD-1), and programmed death ligand 1 (PD-L1) have all demonstrated clinical efficacy. CTLA4 expressed on T cells primarily regulates the extent of the early stages of T-cell activation ⁷² . CTLA4 was the first immune checkpoint receptor to be clinically targeted. The exact mechanism of CTLA4 action is debatable, but its major role is thought to be in the down-modulation of helper T-cell activity and the enhancement of regulatory T-cell immunosuppressive activity ⁷² . Ipilimumab was the first immunotherapeutic to show survival benefit in metastatic melanoma and was approved by the FDA in 2010 for the treatment of advanced melanoma ⁷³ . Therapies directed to other immune checkpoint receptors such as PD-1 and PD-L1 have emerged as promising targets. The PD-1 and B7-1 (also known as CD8) receptors, expressed on activated T cells, are connected by ligands PD-L1 and PD-L2, which are expressed by tumor cells and tumor-infiltrating lymphocytes ^{74– 76} . Tumor PD-L1 expression is common in NSCLC. Indeed, in a study of 436 patients with NSCLC, PD-L1 expression was detectable in tumor cells in 34.5% (88/256) of patients with adenocarcinoma and 34% (61/180) of patients with squamous cell carcinoma ⁷⁷ . PD-1 interaction with both PD-L1 and PD-L2 ligands results in tumor immune recognition and elimination escape by inhibiting T-cell activation. The monoclonal antibody nivolumab directed to the PD-1 receptor has shown activity across NSCLCs with various histologic features. In a randomized trial of 272 patients receiving nivolumab versus docetaxel, the median OS was 9.2 months (95% CI: 7.3–13.3) versus 6.0 months (95% CI: 5.1–7.3), respectively ⁷⁸ . The 1-year OS rate was 42% (95% CI: 34–50) with nivolumab versus 24% (95% CI: 17–31) with docetaxel, while the response rate was 20% with nivolumab versus 9% with docetaxel ( P = 0.008). Nivolumab resulted in a median PFS of 3.5 months, whereas docetaxel demonstrated a median PFS of 2.8 months (HR for death or disease progression: 0.62, 95% CI: 0.47–0.81, P <0.001).

There was a 7% rate of grade 3 or 4 adverse events in the nivolumab group versus 55% in the docetaxel group. In March 2015, nivolumab was approved by the FDA for the management of squamous NSCLC. This was followed by FDA approval of nivolumab for the treatment of non-squamous NSCLC. In October 2015, Borghaei et al. showed that the median OS was 12.2 months (95% CI: 9.7–15.0) among 292 patients in the nivolumab group versus 9.4 months (95% CI: 8.1–10.7) among 290 patients in the docetaxel group (HR for death: 0.73, 96% CI: 0.59–0.89, P = 0.002). At 1 year, the OS rate was 51% with nivolumab versus 39% with docetaxel. At 18 months, the OS rate was 39% (95% CI: 34–45) with nivolumab versus 23% (95% CI: 19–28) with docetaxel ⁷⁹ .

In October 2015, another PD-1 receptor-directed monoclonal antibody, pembrolizumab, was approved by the FDA as a second-line treatment in NSCLC. The clinical efficacy of nivolumab and pembrolizumab was found to be independent of histology. Greater benefit was seen in smokers and in patients with PD-L1-positive expression ⁷⁸ . Although the clinical value of targeting the PD-L1-PD-1 pathway has been established, the POPLAR study evaluated the blockade of PD-L1–B7-1 interactions in addition to PD-L1-PD-1 blockade ⁷⁷ . In a multicenter, open-label, phase II, randomized controlled trial by Fehrenbacher et al., the investigators evaluated the efficacy of atezolizumab versus docetaxel for patients with previously treated NSCLC (POPLAR trial) ⁸⁰ . Atezolizumab is a humanized IgG1 monoclonal anti-PD-L1 antibody which blocks PD-L1-PD-1 and PD-L1-B7-1 interactions, resulting in the restoration of anti-tumor T-cell activity and enhanced T-cell priming. OS in the intention-to-treat population was 12.6 months (95% CI: 9.7–16.4) for atezolizumab versus 9.7 months (95% CI: 8.6–12.0) for docetaxel (HR: 0.73, 95% CI: 0.53–0.99, P = 0.04). The FDA approved atezolizumab in October 2016 for the treatment of patients with metastatic NSCLC whose disease progressed during or following platinum-containing chemotherapy. Several trials comparing anti-PD-1 and anti-PD-L1 agents, individually and in combination with a cytotoxic T-lymphocyte-associated protein 4 inhibitor (ipilimumab), with platinum-based combination regimens as first-line therapy are ongoing ^{81– 83} .

Although targeted therapy agents are directed against tumor cells, they are still associated with toxicities affecting multiple organs. Overall, the most common side effects of tyrosine kinase inhibitors and ALK inhibitors involve the gastrointestinal (43–60%), ophthalmic (55–62%), hematologic (5–10%), and cardiovascular (10%) systems ⁸⁴ . A total of 5% of patients discontinued crizotinib use due to toxicity ⁸⁵ . Liver toxicity is more frequent with gefitinib, and diarrhea and skin toxicity are frequent with afatinib. The most common adverse effects observed with ramucirumab were neutropenia (49%), febrile neutropenia (16%), and fatigue (16%). The number of deaths from adverse events was 5% ⁸⁶ .

The toxicities of immunotherapies in lung cancer are less frequent. The most commonly observed immune-related adverse effects of nivolumab involved the skin, gastrointestinal tract, and kidneys ⁷¹ . All adverse effects including pneumonitis were managed effectively with corticosteroids or infliximab and did not result in mortality. Pembrolizumab has a similar safety profile, but one study demonstrated a <1% incidence of mortality from pneumonitis ⁸⁴ .

Stereotactic ablative radiotherapy (SABR) has become a treatment option for early stage node-negative NSCLC patients who are not operative candidates or refuse surgical options ⁸⁷ . Despite its high tumor control rate and favorable toxicity profile relative to other surgical and non-surgical options, SABR use has been limited primarily to early stage disease ^{87– 90} . Limited data from two pooled studies demonstrated the potential role for SABR in T1–2aN0M0 inoperable NSCLC ⁹¹ . Patients were randomized to SABR (31 patients) versus lobectomy (27 patients). The estimated OS at three years was 95% (95% CI: 85–100) in the SABR group compared with 79% (95% CI: 64–97) in the surgery group (HR: 0.14, 95% CI: 0.017–1.190, log-rank P = 0.037). The SABR group had a recurrence-free survival at three years of 86% (95% CI: 74–100), while the surgery group’s was 80% (95% CI: 65–97). In addition to radiotherapy’s role in stage I NSCLC, new studies have examined its role in combination with chemotherapy in the management of stage IV NSCLC and have shown improved survival when radiotherapy was applied to the primary tumor. From a total of 274 patients in a study by Su et al., 183 had non-oligometastatic disease. Patients with oligometastatic disease who received radiation to primary tumor had better OS ( P <0.001). For non-oligometastatic patients, multivariate analysis showed that receiving ≥63 Gy radiation, having a gross tumor volume of <146 cm ³, having response to chemotherapy, and having stable or increased post-treatment KPS independently predicted better OS ( P = 0.018, P = 0.014, P = 0.014, and P = 0.001, respectively). Among patients with greater primary tumor volume, high radiation dose remained of benefit for OS ⁹² . Radiotherapy has also shown benefit in the setting of CNS involvement. Depending on the volume and number of brain metastases, stereotactic radiation is often preferred to whole-brain radiation therapy (WBRT), when feasible, mainly due to fewer neurological side effects. In patients with EGFR-mutant NSCLC with brain metastasis, a higher response rate is noted with WBRT. Additionally, WBRT improves the duration of intracranial disease control compared to tyrosine kinase inhibitors or stereotactic radiosurgery. Despite these results, many practitioners and patients avoid WBRT to prevent the hair loss, fatigue, and neurocognitive sequelae of WBRT ⁹³ . In patients with oligo-progressive disease on EGFR tyrosine kinase inhibitor monotherapy, local radiotherapy or surgical resection is recommended with continuation of the EGFR tyrosine kinase inhibitor on a case-by-case basis ^{93– 95} .

Various techniques are being investigated to make surgical resection less invasive. Sublobar resection and single-port video-assisted thoracoscopic surgery (VATS) have shown promising results ⁹⁶ . In their study of 121 patients (88% adenocarcinoma, 76% stage I), Hsu et al. used single-port VATS segmentectomies and lobectomies and concluded that resection was both safe and feasible in a multi-institutional retrospective analysis. They identified patient age of 60 years or more, male sex, and tumor size greater than 3 cm as unfavorable preoperative outcome predictors. The results of this study remain to be validated in a randomized controlled prospective study ⁹⁶ . Altorki and colleagues compared patients treated with sublobar resection to patients treated with lobectomy in stage I NSCLC and showed no significant 10-year survival difference among the two groups ⁹⁷ . Additionally, Burdett et al. showed that patients with stage IB, II, or III NSCLC who received chemotherapy following surgery (with or without radiotherapy) had longer survival than those who had surgery without chemotherapy (with or without radiotherapy) ⁹⁷ . The role of surgery has also changed secondary to the implementation of lung cancer screening within and outside the USA.

Early cancer detection using low-dose CT screening showed a 20% reduction in lung cancer mortality in the National Lung Screening Trial (NLST) ^{98, 99} . Major goals of surgical participation in lung cancer screening programs were defined and established in Europe by the European Society of Thoracic Surgeons ¹⁰⁰ . These goals include optimization of the management of screen-detected nodules, reduction of false-positive rates of surgical biopsies, reduction of surgical incision-related trauma, implementation of national or international risk assessment guidelines, implementation of a smoking cessation policy, and active education of primary care physicians towards lung cancer screening programs. Additionally, it is important to emphasize the role of surgical resection in the diagnosis of specific histologic types of lung cancer as per the 2015 World Health Organization (WHO) classification of lung tumors and the eighth edition of the TNM classification. It is emphasized that in tumors such as adenocarcinoma in situ (AIS) and minimally invasive adenocarcinoma (MIA), large cell carcinoma, adenosquamous carcinoma, and pleomorphic carcinoma, the diagnosis cannot be made without complete evaluation of the entire tumor histologically ^{101, 102} . Additionally, although the role of surgery in stage IIIA N2 disease remains controversial, in a meta-analysis of 868 patients from six trials, McElnay et al. showed that the HR comparing patients randomized to surgery after chemotherapy was 1.01 (95% CI: 0.82–1.23, P = 0.954), whereas patients randomized to surgery after chemoradiotherapy demonstrated a HR of 0.87 (95% CI: 0.75–1.01, P = 0.068). The overall HR of all pooled trials was 0.92 (95% CI: 0.81–1.03, P = 0.157) ¹⁰³ . The authors concluded that surgery should be considered as part of the multimodal treatment for patients with resectable lung cancer and ipsilateral mediastinal nodal disease. In the trimodality regimen, a 13% relative improvement in OS was noted with surgical results. Resection for patients with oligometastatic disease represents a relatively new concept in thoracic surgery. In patients with oligometastatic disease, local treatment for distant metastases and curative-intent pulmonary resection may lead to longer survival but requires an optimal method of patient selection. Additional trials are required to determine the ideal treatment plan and long-term follow up in this patient population ^{104– 106} .

The identification of driver mutations has led to a new era of personalized targeted therapy for NSCLC. This has created hope for a disease that carries a high morbidity and mortality. Cutting-edge interrogation of the tumor genome has provided a unique opportunity to investigate potentially targetable molecular aberrations in tumor suppressor genes and oncogenes. Paired with effective targeted therapies, these advances can help improve NSCLC outcomes. However, resistance to targeted therapies and the ongoing demand for better understanding of driver genomic aberrations requires continued and relentless effort. Studies of novel checkpoint inhibitors and numerous trials exploring the role of immunotherapy are ongoing, although there have been promising results with favorable toxicity profiles using immunotherapeutic agents. Combination therapeutics with chemotherapy, immunotherapy, and/or stereotactic ablative therapy need to be studied further.