Perhaps the most important limitation of neoadjuvant clinical trials is the inability to properly stage patients prior to surgical resection; ie, all staging is by definition clinical staging (even if mediastinoscopic lymph node sampling is performed). Inaccurate staging can lead to improper treatments. Positron-emission tomography/computed tomography (PET/CT) and mediastinoscopy should be considered in patients who are potentially resectable, to improve the pre–induction chemotherapy staging.
The published trials have used a variety of different chemotherapy regimens including different platinum analogs (cisplatin or carboplatin(Drug information on carboplatin)). Further studies are needed to determine the optimal chemotherapy regimen and to clarify the role of carboplatin in the curative setting. Many studies have included a mixture of stages, which adds to the difficulty of interpreting the data. Future directions for clinical trials should give consideration to dividing trials into stage IB/II vs IIIA, as outcomes vary between these two groups. This should alleviate some of the difficulties in the analysis and interpretation of the clinical data and its subsequent application to the clinical decision-making process.
Some authors have raised the concern that preoperative chemotherapy may increase the risk of postoperative complications. In 2001, Roberts et al found that neoadjuvant chemotherapy increased the rate of perioperative morbidity.
Matched for age, comorbidities, and pulmonary function, 34 patients who underwent induction chemotherapy were compared with 67 patients who underwent primary surgery. The investigators found an increased rate of life-threatening complications in the neoadjuvant chemotherapy group, including pneumonia, emergency surgery, transfer to the intensive care unit, or intubation (6.0% vs 26.5%, P = .0036), and major complications defined as prolonged hospital stay (19.4% vs 47.1%, P = .0037), reintubation (3.0% vs 17.6%, P = .0093), and tracheostomy (0% vs 11.8%, P = .0042).
One major limitation of this study is the lack of matching for stage between the two groups. Patients in the induction chemotherapy arm had an overall higher stage (2.52 vs 1.55, P < .001). Additionally, findings from this analysis were based on data from a single surgeon at a single institution. Because of these factors, this study should not be used to dictate standard-of-care recommendations. In another retrospective analysis of 106 patients, Venuta et al showed that lobectomy can be safely performed after induction chemotherapy by seasoned surgeons.
Prospective studies have also been done to better assess perioperative risks associated with neoadjuvant chemotherapy. These studies include the larger randomized studies already mentioned (Table 4), which did not demonstrate excessive unexpected morbidity. These studies demonstrated the feasibility and safety of neoadjuvant chemotherapy when used in properly selected patient populations. A criticism has been that these studies pooled all types of pulmonary resections together. When the risks associated with pneumonectomy after induction chemotherapy were assessed, studies supporting and dismissing the safety of neoadjuvant chemotherapy can be found (Table 4).
Mansour et al retrospectively evaluated 306 patients who underwent pneumonectomy from January 1999 to July 2005 and found no significant increase in morbidity or mortality after neoadjuvant chemotherapy. This result is based on a single-site, single-surgeon analysis. Martin et al performed a retrospective analysis of the Memorial Sloan-Kettering Cancer Center thoracic surgical database from 1993 to 1999 and found 479 patients who received induction chemotherapy. This group demonstrated an excess mortality (23.9%) in those who underwent a right pneumonectomy.
In summary, the risks of surgical morbidity and mortality associated with neoadjuvant chemotherapy remain controversial. Although some studies have reported an increased incidence of complications, others have not. Due to the potential excess mortality, a right pneumonectomy after neoadjuvant chemotherapy should only be considered in selected patients and should only be performed by an experienced thoracic surgeon.
Role of Postoperative Radiation Therapy
Phase II studies evaluating the role of induction chemotherapy vary with respect to use of postoperative radiotherapy (PORT). Given concerns over efficacy and safety, this is another controversial area in the treatment of patients with resectable NSCLC. The issue is highlighted by data from 2,128 patients with stage I to III (N2) disease, presented by the PORT Meta-analysis Trialist Group. Overall, PORT was found to be more harmful to patients with resectable lung cancer. A subset analysis suggested that the majority of the detrimental effects occurred in those with stage I/II disease and not in those with N2 disease. Still, this study contained many flaws, such as the inclusion of unpublished data, obsolete radiation techniques now known to be inferior, and inapt staging techniques. More recently, PORT has been evaluated in over 7,400 patients with stage II or III NSCLC, and a benefit was found in those with N2 disease. No benefit was found in those with N0 or N1 disease.
Third-Generation Triplet Combinations
To improve the survival and response rates to preoperative chemotherapy, investigators have explored platinum-based triplet combinations. For example, in a phase II study, 49 patients with stage IIIA (N2) NSCLC received preoperative chemotherapy with cisplatin(Drug information on cisplatin), paclitaxel(Drug information on paclitaxel), and gemcitabine(Drug information on gemcitabine) followed by surgery. Results revealed that 55% of patients underwent complete surgical resection, 73.5% had a reduction in tumor burden, and the 1-year overall survival rate was 85%. Toxicities were associated with the triplet combination, including grade 3/4 neutropenia in 32.7% of patients and thrombocytopenia in 12%.
In another study (a South African phase II trial), the triplet combination of paclitaxel, carboplatin, and gemcitabine was evaluated in patients with operable NSCLC. In the 44 patients with stage IB–IIIA disease, 76.2% had an objective response to therapy, 81% had a complete resection, and the 1-year survival rate was 86%. Toxicities associated with this regimen mainly included grade 3/4 neutropenia (38.6%) and grade 3 thrombocytopenia (25%).
The Spanish Lung Cancer Group trial demonstrated the efficacy of three cycles of cisplatin, gemcitabine, and docetaxel(Drug information on docetaxel) chemotherapy followed by surgery in patients with stage IIIA and IIIB (T4, N0/1) disease, with an overall response rate of 68.9% and 3- and 5-year survival rates of 60.1% and 41.4%, respectively. The incidence of grade 3/4 neutropenia and thrombocytopenia was 62.5% and 25.7%, respectively. These studies showed the feasibility of delivering neoadjuvant platinum-based triplet chemotherapy to operable candidates with NSCLC. The use of these regimens translated into a higher response rate (70% vs 60%) compared with historical controls of platinum doublets. The toxicities seen in these studies were primarily hematologic, and the rates of hematologic toxicities do exceed those seen previously with platinum doublets.
To definitively determine whether triplet combinations improve outcomes compared to doublet regimens, a phase III trial is needed. Participants in this study would need to be followed closely. Before such an investigation is started, however, thought must be given to the question of whether it is truly warranted, as survival with the platinum triplets was not superior to the platinum doublets while the toxicities are increased. Based on the current data, the authors of this article do not believe that three-chemotherapy combinations offer significant advantages that would justify the added risks associated with these regimens.