The optimal management of patients with early-stage NSCLC remains controversial, and the prospect of exposing an otherwise low-risk patient to potentially toxic chemotherapy based on limited prognostic information remains a problem for many clinicians.
The optimal management of patients with early-stage non-small cell lung cancer (NSCLC) remains controversial, and the prospect of exposing an otherwise low-risk patient to potentially toxic chemotherapy based on limited prognostic information remains a problem for many clinicians. New strategies based on proliferation-based expression signatures are under development that may help further refine treatment plans.Survival rates among patients with early-stage NSCLC, which accounts for 80% of all lung cancers, remain poor despite the use of adjuvant chemotherapy.1Surgery has been the mainstay for patients with early NSCLC and, upon pathological staging, recommendations can be made for aggressive or conservative treatments.2,3In stage II and III disease, adjuvant chemotherapy is the standard based on phase III results demonstrating a survival benefit.3,4Pooled meta-analysis of 5 large randomized trials by the Lung Adjuvant Cisplatin Evaluation (LACE), however, was less conclusive for patients with earlier (stage IA and IB) disease, because the benefit of platinum-based adjuvant chemotherapy was most pronounced for patients with stage II and III disease.1,3Studies such as CALGB 9633 also failed to show significant survival benefit in stage IB patients, and thus the optimal management of stage I disease remains controversial.5
Nevertheless, a number of patients with stage I disease may experience recurrence and die from their disease. Fiveyear survival can be as low as 66% and 56% for patients with IA and IB disease, respectively,2and additional stratification factors are needed to identify candidates for aggressive therapy. Results from CALGB 9633 support the role of tumor size as a determinant of survival in the IB subset, suggesting that adjuvant chemotherapy for patients with stage IB tumors >4.0 cm could be considered.5However, the prospect of exposing lowrisk patients to potentially toxic chemotherapy on the basis of limited prognostic data remains a conundrum for many oncologists.One of the more promising areas of research has been to identify RNA expression signatures associated with poor outcome, such as a 14-gene expression profile that has been used to stratify patients with nonsquamous, resectable NSCLC into high- and low-risk subgroups.6More recently, a study has utilized a signature focused on cell cycle progression (CCP) genes, which may provide assessment of the tumor’s proliferative capacity.4This study focused on the subgroups of patients with stage I and stage II lung cancer in an effort to identify, using the CCP score, both lowrisk patients, who might be spared the use of potentially harmful adjuvant chemotherapy, and higher-risk patients, who might benefit from aggressive treatment.4The study demonstrated that the CCP signature was an independent predictor of lung cancerrelated death in patients with stage I and stage II lung adenocarcinoma, and that patients could be stratified into high-, medium-, and low-risk subgroups, with the lowest CCP score tercile indicative of higher survival.
The CCP score was also able to predict outcomes using stage I and II formalinfixed paraffin-embedded specimens.4In a separate analysis of stage I samples, among the 3 data sets examined, cancerspecific survival at 5 years ranged from 75% to 79% for patients with a CCP score higher than the median, compared with 90% for patients with a lower CCP score. It also showed a significantly greater adjuvant treatment benefit for patients with higher CCP scores compared with lower scores (P = .0060), and this significance was retained after adjusting for clinical variables (P = .0020).4
In a presentation from the 15th World Conference on Lung Cancer, this CCP signature was tested in 2 populations of patients with NSCLC adenocarcinoma who had received resection with no neoadjuvant, adjuvant chemotherapy, or radiotherapy within 12 weeks of surgery. The results showed CCP score to be predictive of 5-year lung cancer mortality after adjusting for stage and other variables, and the prognostic score (PS) was significantly more predictive than stage alone. As such, the low PS group had significantly better 5-year lung cancer mortality than did the high PS group (See Figure).7At the same conference, a presentation by Kim et al reviewed an integrated prognostic model combining CCP score with pathological staging to maximize the prognostic value of both methods in stage I and II patients.8The modeling set included 495 patients from 2 cohorts, and the outcome was 5-year lung cancerspecific survival.8The results showed that the HRs for CCP (1.39) and pathological stage (1.69) were consistent in each cohort for untreated patients, and a risk curve for 5-year lung cancer mortality that was developed.8The results of this study showed that the combination of CCP score with pathological stage was a more effective predictor of cancer-specific death than pathological stage alone.Recognizing the need for additional determinants to help make treatment decisions, Howard (Jack) West, MD, from the Swedish Cancer Institute, noted that the overall recommendation for adjuvant therapy is vague, and he cautioned that, inevitably, some patients will receive potentially toxic chemotherapy that may not be beneficial. West also noted that “extrapolating the highly selected populations from the adjuvant trials for the real-world experience is a challenge.” In early-stage disease, “There are a fair number of patients who fall in the range where you could justifiably go either way.” Highlighting the need for additional data, he concluded that “While the evidence is still limited… it’s compelling,” noting that clinicians may use such assessment tools combined with anatomic stage, as well as host factors, issues of life, kidney function, and the patient’s overall desire for additional therapy, when making treatment decisions for early-stage NSCLC.
Prognostic testing based on the use of such proliferation-based expression signatures is currently available; myPlan™ Lung Cancer is an RNA expression system based on 31 CCP genes, and 15 control genes, used to assess how fast lung cancer cells are replicating.4The myPlan prognostic score combines CCP score and pathologic stage to predict 5-year lung cancer-specific survival.8The assay, marketed by Myriad Genetics, is now commercially available via an early-access program.