The article by Calhoun and colleagues, published in this issue of ONCOLOGY, is a timely review of one of the more controversial questions in thoracic oncology: whether or not patients with stage IB non–small-cell lung cancer (NSCLC) should receive adjuvant chemotherapy.
Lack of Survival Advantage
Within the past 5 years, results have become available from six large randomized trials evaluating the role of adjuvant platinum-based chemotherapy in patients with completely resected NSCLC. Of these, three showed improvement in overall 5-year survival with adjuvant chemotherapy (International Adjuvant Lung Cancer Trial [IALT], National Cancer Insitute of Canada’s JBR.10, Adjuvant Navelbine International Trialist Association [ANITA]), and three others showed no benefit (Big Lung Trial [BLT], Adjuvant Lung Cancer Project Italy [ALPI], Cancer and Leukemia Group B [CALGB] 9633).
CALGB 9633 initially reported improved survival—a finding that disappeared with longer follow-up. An additional study performed in Japan reported improved survival in stage I patients with an oral fluorouracil(Drug information on fluorouracil) analog. All seven studies included patients with stage IB NSCLC. Furthermore, results of two recent meta-analyses have confirmed an absolute improvement in 5-year survival of about 5% with the use of adjuvant chemotherapy.
So where does the controversy lie? As well summarized by Calhoun and colleagues, subset analyses of the three “positive” platinum-based trials have demonstrated no significant survival advantage for patients with stage IB disease. As none of the trials were powered to detect survival differences within individual stages, however, the absence of positive results is not necessarily evidence of futility. As the rationale for adjuvant chemotherapy is the neutralization of micrometastatic disease, it follows that stages where the micrometastatic burden is greater (stages II and III) will be likeliest to be affected by systemic adjuvant treatment. Nevertheless, since 35% to 60% of patients with stage IB NSCLC will eventually die—most due to metastatic disease—it is reasonable to assume a potential role for adjuvant chemotherapy in this stage group.
CALGB 9633 was the only trial that was specifically designed and powered to evaluate the effect of adjuvant chemotherapy in stage IB patients. Unfortunately, this study was terminated early because of slow accrual, having reached only 69% of its target goal. Nevertheless, at interim analysis, an absolute survival advantage of 12% was seen at 4 years. With subsequent maturation of the data, no survival benefit was apparent at 5 years, yet there remained a statistically significant improvement in disease-free survival in the adjuvant chemotherapy arm (hazard ratio = 0.74; 90% 2-sided confidence interval = 0.57–0.96; P = .027). At last review in 2006, only 131 of 150 deaths required for final analysis had been observed. It may therefore be premature to discount the effect of chemotherapy for resected stage IB NSCLC until a final analysis of this trial has been performed.
In addition, the Japanese study of patients with stage I NSCLC treated with adjuvant uracil/tegafur (UFT) showed that survival benefit was limited to patients with stage IB. However, given the apparent geographic differences in the biology of NSCLC, these results require further validation.
The Heterogeneity Issue
Calhoun and colleagues also draw attention to the important issue of heterogeneity within the stage IB subset. Recognizing the influence of tumor diameter on survival, the next iteration of the American Joint Committee on Cancer (AJCC) staging system (7th edition) will reclassify tumors > 5 to 7 cm as stage IIA, and those > 7 cm as IIB. This will lead to better stratification of patients enrolling in future clinical trials, but will also have implications regarding how we interpret the results of previous phase III studies. Trials that enrolled stage IB patients in the past included significant numbers of patients with tumors > 5 cm, which will be classified as stage II under the new staging system.
Thus, stage IB may soon represent a subset of patients who have significantly better survival than the current stage designation implies. It is even less likely, therefore, that adjuvant systemic therapy would benefit this population as defined by the revised staging system. In support of this premise, subset analysis of CALGB 9633 showed that survival advantage was limited to IB patients with tumors > 4 cm. Additionally, it is well known that factors other than size also influence tumor recurrence. As Calhoun and colleagues rightly point out, biologic parameters such as tumor histology, lymphovascular invasion, tumor differentiation, and molecular profile may have a significant impact on survival. None of these factors are taken into account in the current staging system.
We will soon enter an era of biologic staging of cancer, where stage designation will be determined not only by the physical and geographic characteristics of the tumor, but also by its genomic makeup. Two of the positive randomized adjuvant trials included analysis of a limited number of molecular markers. The IALT study showed that patients with p27Kip1-negative tumors benefited from chemotherapy after surgery, whereas those who were p27Kip1-positive did not. Similarly, the JBR.10 study demonstrated that patients with wild-type ras proto-oncogene benefited from adjuvant chemotherapy, but those with mutated ras did not. Single oncogenic markers, however, are unlikely to be useful for predicting prognosis in an individual patient because of the great variation of expression of any one marker in any given tumor.
A more comprehensive analysis of molecular risk stratification was performed by Potti and colleagues. These investigators analyzed gene-expression arrays in patients with resected NSCLC and showed that prediction of tumor recurrence and survival using a gene expression–based model (the so-called lung metagene model) was significantly more accurate than using standard clinical models that included stage, tumor size, age, sex, histology, and smoking status. Not only has the lung metagene model been validated in two separate cohorts of resected lung cancer patients, but it has been used to define a subset of patients with stage IA NSCLC who have significantly worse 4-year survival than their counterparts (< 10% vs 70%).
Adjuvant chemotherapy is not usually recommended for stage IA NSCLC, and yet the metagene model was able to define a group at very high risk for recurrence who perhaps would have benefited from systemic treatment. This implies that stage alone may not be the best determinant for deciding whether patients should receive adjuvant treatment. A prospective trial is planned, which will evaluate the effectiveness of a metagene-based algorithm for selecting patients with resected early-stage NSCLC who will benefit from adjuvant chemotherapy.
In summary, the current data do not support routine use of adjuvant chemotherapy for stage IB NSCLC. While no trials of platinum-based therapy have shown an overall survival advantage, a single study has demonstrated improved disease-free survival and overall survival in patients with tumors > 4 cm. If a benefit for stage IB disease does exist, it will probably be most evident in patients with larger tumors, most of which will be classified as stage IIA or IIB by the new staging system. Molecular-based risk assessment complements clinical staging, and in the future, a staging system based on tumor, nodes, metastases, and biology (TNMB) might refine our ability to predict prognosis and determine appropriate adjunctive therapies.
—David Rice, MB, BCh, FRCSI