Surgery in Stage I and Stage II Non–Small Cell Lung Cancer

Surgical resection is the treatment of choice for patients with localized NSCLC whose pulmonary function and general medical condition permit this to be done safely. Surgical options are discussed in greater detail in Chapter 2 , on the surgical staging of lung cancer.

Use of Postoperative Radiotherapy In Stage I and Stage II Non–Small Cell Lung Cancer

Postoperative radiotherapy to reduce risk of local recurrence generally is used if surgical margins are positive for tumor. In patients with clear surgical margins, randomized studies comparing outcomes with postoperative irradiation and with observation have failed to demonstrate an advantage for radiotherapy. In fact, postoperative radiotherapy is associated with an increased risk of death, with absolute survival rates of 55% at 2 years without radiotherapy and 48% with radiotherapy. Negative impact on survival is greatest in stage I disease and less in stage II, with no negative (or positive) impact on survival in patients with stage III disease. Local tumor recurrence rates were reduced by 24% with postoperative radiotherapy, suggesting a potential positive effect on tumor control but negative effect on survival owing to radiotherapy toxicity. Hence, for a majority of patients with stage I or II NSCLC, the current standard of care is not to administer postoperative radiotherapy, whereas it is frequently administered in patients after resection of stage III disease and in other patients in whom clear surgical margins could not be achieved, to reduce risk of local mediastinal recurrence.

Radiotherapy as Definitive Treatment for Medically Inoperable Stage I and Stage II Non–Small Cell Lung Cancer

A high proportion of patients with lung cancer have comorbid conditions such as chronic obstructive lung disease or coronary artery disease. In some patients, these comorbid illnesses are severe enough to preclude surgery. Although even very elderly patients may benefit from surgery if they are in good health, older patients also are more likely to have comorbid conditions that make surgery inadvisable. Radiotherapy alone may lead to long-term control in some patients with NSCLC, with overall 5-year survival rates ranging from 5% to 42% for different studies and a median 5-year survival rate across studies of approximately 15%. To correct for the effect of comorbidity, cause-specific survival also has been assessed, and in most studies, cause-specific 5-year survival rates generally have been 10% to 20% higher than overall survival rates, with a median survival across studies of approximately 31 months. Although no randomized comparisons of definitive radiotherapy and surgery for patients with stage I and stage II NSCLC have been conducted, median survival times for early-stage NSCLC across broad populations (in which a majority of the patients had undergone surgery) are longer than the cause-specific survival noted above with radiotherapy. Although comorbid conditions could account for differences in overall survival between surgery and radiotherapy, they would be less likely to have an impact on the cause-specific survival, suggesting better local tumor control with surgery. Nevertheless, radiotherapy is a useful option for patients who are not surgical candidates.

Stereotactic Radiotherapy for Stage I and Stage II Non–Small Cell Lung Cancer

Recent studies suggest that stereotactic radiotherapy may be of value in medically inoperable patients without mediastinal node involvement. This modality typically entails delivery of radiation to a localized area in a single large treatment or as three to four fractions over 1 to 2 weeks, with a high dose each day, for a total radiation dose of approximately 45 to 60 Gy. Local control rates of more than 80% have been achieved with small peripheral tumors and approximately 50% with central tumors. Severe toxicity is more likely with large tumors or with centrally located tumors, whereas stereotactic radiotherapy generally is well tolerated for smaller peripheral tumors. A randomized trial comparing stereotactic radiotherapy and surgery in appropriate candidates is ongoing.

Postoperative Adjuvant Chemotherapy in Stage I to Stage IIIA Non–Small Cell Lung Cancer

Patients who have undergone resection of NSCLC with curative intent often will demonstrate distant metastases a few months to a few years after surgery. Typically, large numbers of circulating tumor cells can be detected in the blood of patients with invasive malignancies, but most of these tumor cells may die or else remain quiescent after lodging in distant tissues and never form detectable metastases. However, some of these surviving tumor cells also may begin to proliferate, eventually manifesting as metastases. Although chemotherapy cannot permanently eradicate tumors large enough to be seen radiologically, it may in some cases be able to eradicate micrometastases.

Recent randomized trials have demonstrated that postoperative adjuvant cisplatin-based chemotherapy may improve survival in patients after resection for stage II or III NSCLC, particularly in those with good performance status, with a 5-year absolute survival benefit of 5.4% from chemotherapy. The benefit varied with stage, with a possible harmful effect seen in patients with stage IA disease (hazard ratio [HR] = 1.40; 95% confidence interval [CI], 0.95–2.06), equivocal outcome in those with stage IB disease (HR = 0.93; 95% CI, 0.78–1.10), but evidence of benefit in those with stage II disease (HR = 0.83; 95% CI, 0.73–0.95) and in stage III (HR = 0.83; 95% CI, 0.72–0.94). Because prognosis worsens with increasing tumor size and with pleural or vascular invasion by tumor, adjuvant chemotherapy may in some cases also be offered to patients with stage IB disease whose tumors are larger or exhibit pleural or vascular invasion, despite the fact that no randomized clinical trials assessing benefit in this subpopulation have yet been conducted. Studies in Japanese patients with NSCLC suggested that the fluorinated pyrimidine-uracil compound UFT may be of value for adjuvant therapy in patients with resected stage I disease, but this agent has not yet gained acceptance in North America or Europe, and its efficacy has been questioned.

Typically, three to four cycles of adjuvant chemotherapy are administered over 9 to 12 weeks. Some clinicians substitute carboplatin for cisplatin, because it is less toxic. In advanced NSCLC, however, carboplatin is not quite as effective as cisplatin, and proof of benefit of carboplatin regimens for adjuvant therapy is lacking, because only limited studies have been done. With respect to the drug added to cisplatin or carboplatin, vinorelbine, another vinca alkaloid, and etoposide each have been used with some success. Taxanes such as paclitaxel and docetaxel are at least as effective as vinca alkaloids and etoposide against advanced metastatic NSCLC and also are frequently used in the adjuvant setting despite a lack of definitive testing of these agents for such therapy.

Preoperative (“Neoadjuvant”) Chemotherapy in Non–Small Cell Lung Cancer

Chemotherapy may also be administered before surgery. Such neoadjuvant chemotherapy has several potential advantages over postoperative adjuvant chemotherapy, including improved drug tolerability, improved drug delivery, preoperative downstaging (which may improve resectability and prognosis), decreased perioperative tumor seeding, and early control of micrometastases. In some studies, neoadjuvant chemotherapy and chemoradiotherapy have decreased tumor size sufficiently to render unresectable stage IIIB disease in patients with NSCLC potentially resectable, and in some small randomized studies in patients with operable stage I, II, or IIIA NSCLC, neoadjuvant chemotherapy significantly improved survival. A larger study in patients with operable stage IB to stage IIIA NSCLC found a trend toward improved progression-free and overall survival, and in a meta-analysis of six randomized trials, administration of neoadjuvant chemotherapy in stage I to stage IIIA NSCLC was associated with an HR of 0.66 (95% CI, 0.48–0.93) compared with surgery alone. Neoadjuvant chemotherapy is now considered an option as a standard of care for selected patients with operable stage III NSCLC. In NSCLC meta-analyses, the HR of 0.66 with neoadjuvant chemotherapy compared favorably with the HR of 0.84 (95% CI, 0.78–0.89) seen in 19 trials of postoperative chemotherapy, although studies comparing neoadjuvant chemotherapy to postoperative chemotherapy in NSCLC have not yet been done.

Currently, neoadjuvant chemotherapy often is used in patients in whom preoperative staging reveals low-bulk involvement of subcarinal or ipsilateral mediastinal nodes. Patients with bulky disease involving ipsilateral mediastinal nodes or with known involvement of contralateral mediastinal nodes or of more than one mediastinal lymph node station generally are not considered to be candidates for surgery because of the high risk of later development of distant metastases despite adjuvant chemotherapy. Neoadjuvant chemotherapy also is occasionally used in patients with bulky disease in stage II NSCLC, particularly if tumor shrinkage in response to chemotherapy potentially may offer the possibility for less extensive surgery (e.g., by permitting lobectomy in a situation in which pneumonectomy would otherwise be required).

Radiotherapy for Inoperable Stage III Non–Small Cell Lung Cancer

Radiotherapy alone may be effective in a small proportion of patients with inoperable stage IIIA NSCLC and may also cure some patients with stage IIIB disease, in the absence of a malignant pleural effusion (i.e., in “dry IIIB” as opposed to “wet IIIB” disease), with long-term survival rates for patients with stage IIIA and stage IIIB NSCLC of approximately 5% to 10%. The typical radiotherapy dosing schedule is once a day, 5 days a week, for 6 to 6.4 weeks, to deliver a total dose of 60 to 64 Gy. This schedule takes advantage of the fact that normal tissues are more efficient than tumor tissues at repair of radiation damage, to maximize the therapeutic index when it is necessary to include substantial volumes of normal tissue with the radiation.

As a rule, the greater the bulk of tumor, the lower the probability of cure with radiotherapy, because more bulky disease is associated with increased risk of later distant metastases and also is associated with a higher rate of local failure. Involvement of supraclavicular lymph nodes also increases the risk of eventual failure. Outcomes with radiotherapy are somewhat better for squamous cell carcinoma than for other tumor types.

The location of the primary tumor also may be relevant. If a patient has a peripheral lower lobe primary tumor and nodal involvement, then the size of the radiotherapy field and the amount of lung destroyed by the radiotherapy will be relatively large, whereas the size of the radiotherapy field is much smaller and the amount of radiation damage to lung is much less if the patient has a centrally placed upper lobe tumor, and the probability of survival is higher with inoperable upper lobe tumors than with inoperable tumors in other locations.

Radiation has to travel through the lung to reach the tumor and subsequently passes out of the far side of the lung; in the process, this energy destroys lung tissue in its path. Scatter to more distant lung also may occur, and the resultant exposure may be sufficient to cause extensive radiation pneumonia, which may respond to steroids but also can prove fatal in some instances. Other organs and structures at risk for damage from the radiation include skin, esophagus, heart, and spinal cord. Radiation esophagitis may be severe enough to necessitate temporary placement of a feeding tube and may result in later scarring with consequent esophageal stricture formation requiring periodic esophageal dilatation. An excessive radiation dose to the spinal cord may damage blood vessels, causing irreversible transverse myelitis with paraplegia. Great care is taken to limit the dose to the spinal cord, but use of the angles required to accomplish this often results in higher doses to the esophagus, with increased esophagitis.

Over the past several years, progressive improvements in radiotherapy technology have permitted delivery of higher, more effective radiotherapy doses with reduced toxicity. Orthovoltage equipment gave way to cobalt 60 equipment, which in turn was replaced by linear accelerators. Advances in radiotherapy planning methods ultimately permitted further refinement of linear accelerator techniques, first with three-dimensional conformal radiotherapy (3DCR), then with intensity-modulated radiotherapy (IMRT), and then with IMRT with mechanisms to correct for movement of the tumor with respiration. In IMRT, a relatively large number of radiotherapy beamlets with differing dose intensities are directed at the tumor from different angles. With IMRT, a very large volume of low-dose radiation is delivered with the entrance and exit of these many beamlets, but in comparison with 3DCR, a smaller volume of lung receives 20 Gy (considered to be the threshold for development of radiation pneumonitis), and the observed rate of treatment-related pneumonitis is significantly lower with IMRT than with 3DCR.

Recently, a growing experience with proton beam radiotherapy (PBR) has been the subject of study. PBR has the advantage that it deposits all of its radiation energy at the tumor; accordingly, although toxicity to tissues is incurred along the path of radiation into the tumor, toxicity resulting from radiation exiting the far side of the tumor is greatly reduced. Preliminary data suggest that the observed clinical reduction in toxicity with PBR matches that anticipated in theoretical analyses.

Chemoradiotherapy for Inoperable Stage III Non–Small Cell Lung Cancer

Randomized trials showed that addition of chemotherapy to radiotherapy improved the probability of long-term control in locally advanced, inoperable NSCLC. The initial studies involved administration of a few cycles of platinum-based chemotherapy, followed by approximately 6 weeks of radiotherapy. Subsequent studies showed that administration of low-dose chemotherapy with once-daily or once-weekly dosing during radiotherapy also improved survival by improving control of the local tumor. Still later randomized studies demonstrated that administration of chemotherapy concurrently with radiotherapy is superior to sequential administration of chemotherapy followed by radiotherapy, and this is now the standard of care for patients with stage IIIA and “dry” stage IIIB NSCLC.

Chemotherapy may help improve outcome by several different mechanisms. In addition to directly adding to tumor cell kill, cisplatin and carboplatin can act as free electron acceptors—thereby taking the place of oxygen and prolonging the half-life of DNA-damaging hydroxyl radicals produced by radiation—and can inhibit repair of radiation-induced DNA damage. Both the platinums and a variety of other agents such as the taxanes and vinca alkaloids also can increase the proportion of tumor cells that are in the radiation-sensitive G ₂ /M phase of the cell cycle, and various other mechanisms also may be at play. Although the addition of chemotherapy to radiation therapy does increase toxicity to the esophagus, this toxicity is manageable.

Palliative Radiotherapy for Advanced Non–Small Cell Lung Cancer

Stage IV and “wet” stage IIIB NSCLC—so-called advanced NSCLC—is incurable, so the objective of therapy is symptom control. Radiotherapy (with a typical dose of 30 Gy given in 10 fractions over 2 weeks or 45 Gy given in 15 fractions over 3 weeks) may be useful for reducing the frequency or severity of a variety of cancer-induced symptoms, including pain, cough, hemoptysis, and dyspnea. It also can reduce the risk of several complications of metastases, including pathologic fractures from bone metastases, paralysis from spinal cord compression, and neurologic complications of brain metastases. Irradiation also may open up airways or esophagus obstructed by tumor, thereby facilitating breathing and eating.

First-Line Chemotherapy for Advanced Non–Small Cell Lung Cancer

In a recent meta-analysis of randomized studies, the median survival time for patients with advanced NSCLC who did not undergo chemotherapy was 4.5 months and the 1-year overall survival rate was 20%; in patients who received chemotherapy, median survival was 6.0 months and the 1-year survival rate was 29% (HR = 0.75; 95% CI, 0.67–0.84; P < .0001). The beneficial effects of chemotherapy were observed irrespective of age, gender, histologic features, type of cytotoxic drug (although long-term alkylating agents were associated with a detrimental effect in an earlier meta-analysis and were excluded from this updated report), or baseline performance status (although more than 75% of the patients had a performance status of 0 or 1). Quality of life either was improved after chemotherapy or was comparable to that seen in the best supportive care group.

Two-drug combinations are superior to single-agent chemotherapy in advanced NSCLC, although such regimens also are associated with somewhat greater toxicity. Platinums (cisplatin or carboplatin) form the backbone of most NSCLC combination regimens, and meta-analyses suggest that cisplatin is slightly more effective than carboplatin, although it also is more toxic. Docetaxel, paclitaxel, vinorelbine, and gemcitabine all give similar results when added to a platinum, although one study suggested that docetaxel may be somewhat superior to vinorelbine when added to cisplatin, with a median overall survival of 11.3 months versus 10.1 months (HR = 1.183; 97.2% CI, 0.989–1.416; P = .044) and better quality-of-life scores. In one study, progression-free survival was slightly longer with the use of gemcitabine than with taxane regimens. However, all of these agents generally are looked upon as being equivalent in efficacy when added to a platinum, yielding response rates in the range of 17% to 32% and a median survival of 7.8 to 11.3 months. Older agents such as etoposide, vinblastine, and vindesine are somewhat less effective. The topoisomerase I inhibitor irinotecan also may be helpful when combined with a platinum in advanced disease.

The combination of cisplatin with the multitargeted antifolate pemetrexed was superior to the combination of gemcitabine with cisplatin in patients with adenocarcinoma and large cell carcinoma, whereas the gemcitabine combination was superior to the pemetrexed combination in patients with squamous cell carcinomas. Adding a third chemotherapy agent to a two-drug platinum-based regimen results in increased response rate but does not improve survival.

Typically, four to six cycles of chemotherapy are administered to patients who are responding. Chemotherapy is not continued for longer periods because the first few cycles of chemotherapy appear to be the most effective ones. Moreover, randomized trials have not demonstrated any benefit to proceeding with chemotherapy beyond the initial four to six cycles, although recent studies of maintenance pemetrexed after initial chemotherapy with other agents suggest a possible benefit in patients with adenocarcinoma. In some patients, rapid tumor regrowth occurs when therapy is discontinued; in others, regrowth may not occur until several months later.

Chemotherapy in Previously Treated Patients with Advanced Non–Small Cell Lung Cancer

Tumors that have previously been exposed to chemotherapeutic agents are less likely than chemo-naive tumors to respond to subsequent chemotherapy. Nevertheless, randomized studies comparing docetaxel with best supportive care in previously treated patients demonstrated that docetaxel provided a benefit, with an improvement in median life expectancy from 4.6 months to 7 months, as well as an improvement in average quality of life. Subsequent studies demonstrated that pemetrexed was as effective as docetaxel in the second-line setting but was less toxic. Gemcitabine also may be of some value in this setting, whereas vinorelbine appears to be ineffective.

Targeted Therapies in Non–Small Cell Lung Cancer

Targeted therapies are based on the use of drugs or other agents that are directed specifically against a cell growth receptor or its ligands, signaling pathway components, or other effectors of function. Several such agents currently are under investigation for the treatment of lung cancer. The monoclonal antibody bevacizumab (Avastin) binds to and inactivates vascular endothelial growth factor (VEGF), thereby antagonizing the new vessel formation (angiogenesis) required by tumors for growth. Randomized trials confirmed that addition of bevacizumab to carboplatin plus paclitaxel increased both response rate (from 15% with chemotherapy alone to 35% with chemotherapy plus bevacizumab) and median survival (10.3 months and 12.3 months, respectively). Preliminary assessments suggest that vandetanib, a small molecule that inhibits both the VEGF receptor and the epidermal growth factor receptor (EGFR), also may prolong progression-free survival when added to chemotherapy regimens for treatment of NSCLC.

In a comparison of outcomes achieved with the EGFR tyrosine kinase inhibitor (TKI) erlotinib and with best supportive care, erlotinib treatment significantly prolonged survival of patients with NSCLC previously treated with chemotherapy, from a median of 4.7 months with best supportive care to 6.7 months with erlotinib, with improvement also noted in average quality of life. Treatment with erlotinib is now considered a standard of care in this setting. This agent is most effective in patients with EGFR-activating mutations or high EGFR gene copy number, and with K-ras wild type.

The related EGFR TKI gefitinib also is effective against NSCLC in selected patients, but improvement in survival over that observed with best supportive care failed to achieve statistical significance ( P = .087). Gefitinib is no longer used in North America but is widely used in Asia. Of interest, however, gefitinib recently was found to be comparable in efficacy to docetaxel in the second-line setting, so this agent may be assessed further.

Female patients with lung adenocarcinomas who have never smoked and are East Asian are particularly likely to have EGFR mutations and also are particularly likely to benefit from treatment with an EGFR TKI, and in a randomized trial, gefitinib was superior to standard chemotherapy in chemotherapy-naive Asian never-smokers. Addition of an EGFR TKI to first-line chemotherapy did not result in improved survival, except in patients who had never smoked and in fact was associated with decreased survival in patients whose tumors had K-ras mutations. For reasons that are unclear, administration of gefitinib after completion of chemoradiotherapy for locally advanced NSCLC also worsened overall survival despite having minimal impact on progression-free survival, so this protocol is not currently recommended in this setting.

Although the EGFR TKIs do not improve outcome when added to first-line chemotherapy regimens in advanced NSCLC, addition of the anti-EGFR monoclonal antibody cetuximab (C-225 [Erbitux]) resulted in modest improvement of median overall survival, from 10.1 months in the control group to 11.3 months in the cetuximab-treated patients (HR = 0.871; 95% CI, 0.762–0.996; P = .0441). Additional studies combining cetuximab with chemotherapy plus bevacizumab are under way.

The Ras/Raf/VEGFR inhibitor sorafenib and the proteasome inhibitor bortezomib also appear to have modest activity in NSCLC but do not yet have an established role in clinical practice. Because amplification of c-Met is a potential cause of acquired resistance to EGFR TKIs, clinical studies of use of c-Met inhibitors in NSCLC have recently been launched. Earlier trials of matrix metalloproteinase inhibitors yielded negative results.

Management of the Patient with a Solitary Metastasis from Non–Small Cell Lung Cancer

Although widely metastatic NSCLC generally is considered to be incurable, patients with solitary metastases to one or two sites may potentially achieve long-term survival if such disease is resected. The largest experience with metastasis resection in NSCLC involves solitary brain metastases and, to a lesser extent, adrenal gland metastases. Randomized trials have demonstrated that brain metastasis resection is superior to cranial irradiation alone and that survival is further improved by administering whole-brain radiotherapy after the brain metastasis resection. Aggressive local therapy for solitary bone metastases generally does not result in long-term disease control.