Preoperative neoadjuvant therapy in non-small-cell lung cancer: open season?

Neoadjuvant therapy refers to chemotherapy or radiation administered before surgery when surgery is intended as the definitive local/regional treatment (7). The objectives of neoadjuvant therapy are to enable the resection of an otherwise unresectable neoplasm, to reduce the risk of local or of distant recurrences, and to increase the survival rate. A review of the recent literature indicates that well over 500 patients with Stage IHA lung cancer have been treated with neoadjuvant therapy (2). In addition to these patients, an untold number of patients throughout the country have been treated "off protocol" (ie, not as part of an organized clinical trial), a fact which reflects the enthusiasm for this treatment, not only among university-based oncologists but also among community-based oncologists. However, it is difficult to interpret the data because patients in these various studies are not comparable. Preoperative staging procedures have not been consistent. Some of the criteria have included disease localized to the thorax but deemed unresectable. Others have included patients with either stage IIIA or mB disease. In addition, all these trials have been phase II trials. Drawing conclusions regarding survival from phase II trials is very hazardous, since one cannot control for the impact of factors other than the effect of therapy, such as weight loss, performance status, and physician selection — any or all of which may be responsible for seemingly improved survival. Nevertheless, the data from these phase II studies are most provocative. The clinical response rates to the neoadjuvant therapy of 50% to 70% and the histologic complete response rate of 20% have led many to assume that neoadjuvant therapy will result in improved survival. This assumption may or may not be true. It is apparently not true in the cases of head and neck cancer (3,4) and esophageal cancer (5). Therefore, one cannot automatically assume that clinical and histologic responses to neoadjuvant therapy will translate into improved survival. Unfortunately, patients are told that surgery is the only curative modality in lung cancer. The statement is true as it relates to patients with T1N0 and T2N1 disease, to certain patients with T3N0 disease, and to a very small subset of patients with N2 disease. However, patients with more advanced lung cancer confined to the thorax are often told to "have surgery in case it might help you — it's your only chance." This advice hardly seems justified by the evidence. Perhaps neoadjuvant therapy is popular because it involves physicians from all of the treatment modalities. Placing these considerations aside, does the experience with the more than 500 patients who have received neoadjuvant therapy support the conclusion that surgery improves their cure rate or quality of life? There is one fact that is crystal clean surgery in lung cancer is almost never palliative. Also, with the current state of the art, subtotal resection or "debulking" is never indicated. Is it logical to assume, based on experience with other tumors, that surgery actually improves the cure rate for patients receiving neoadjuvant therapy compared with that for patients receiving either chemotherapy or radiation therapy alone (or a combination of the two) but no surgery? Certainly, the response rates of 60% to 70% are promising. However, the median duration of survival for the resected patients in all of these phase II studies ranges from 16 to 25 months. This survival period is not much different from that experienced by historical controls treated with combined radiation dierapy and chemotherapy but without surgery (<5). Is there a subset of patients who may benefit from neoadjuvant therapy? While the paper by Weiden and Piantadosi (7) in this issue of the Journal does not document an increased duration of survival for patients with histologic complete responses, other investigators have reported improved survival in these patients (2). It must be emphasized that in phase II trials survival analyses are subject to considerable bias (due to variability of patients included and treatments employed) and are simply not reliable. What is clear from these phase II trials is that neoadjuvant therapy does alter patterns of recurrence. Local recurrences are dramatically diminished in these patients; however, systemic recurrences, especially brain recurrences, greatly offset this benefit Therefore, the data and even the logic cannot lead one to assume that surgery adds any benefit to chemotherapy and/or radiation therapy in otherwise surgically incurable patients. In order to answer these questions, prospective randomized phase III trials must be done. The Southwest Oncology Group is attempting to obtain funding for a randomized study comparing (a) neoadjuvant etoposide and cisplatin combined with radiotherapy, followed by surgery and additional chemotherapy, to (b) radiation therapy alone. Such phase III randomized trials will clearly address the issue of the role of surgery in patients who have responded to neoadjuvant therapy. Unfortunately, there are few NCI funds targeted for non-small-cell lung cancer neoadjuvant trials, and, therefore, it is unlikely that such trials will move forward at an acceptable pace. It is a great pity that the number one cancer killer in men and women in the United States is allocated so few resources at the federal level.