Lung cancer assay predicted survival after surgery for early disease

Kratz JR. Lancet. 2012;doi:10.1016/S0140-6736(11)61941-7.

A lung cancer assay demonstrated an ability to accurately predict which patients will be cured by surgery and those who at a heightened risk for death within 5 years of the operation, according to study results published in The Lancet.

The 14-gene assay — initially developed at the University of California, San Francisco, by researchers working with a cohort of 361 patients with nonsquamous non–small cell lung cancer — uses quantitative polymerase chain reaction analysis of formalin-fixed, paraffin-embedded tissues.

The test then was corroborated in two independent clinical trials. One was a masked study using tissue sample analyses from 433 patients with early-stage lung cancer at Kaiser Permanente Hospitals in Northern California. The other study involved 1,006 patients with early-stage lung cancer from the China Clinical Trials Consortium (CCTC).

Eligibility to enter the Kaiser Permanente study was based upon complete resection of American Joint Commission on Cancer stage I nonsquamous NSCLC by clinical and pathological staging at a Kaiser Permanente facility from 1998 to 2005.

Patients were eligible to be included in the CCTC study if they had undergone an attempt at remedial resection for stage I to stage III nonsquamous-cell NSCLC at First Affiliated Hospital of Guangzhou Medical College (Guangdong, China), Sun Yat-sen University Cancer Centre in Guangzhou (Guangdong, China) or Shanghai Pulmonary Hospital (Shanghai, China) from 2000 to 2008.

Analyzing the activity of 14 genes in tumor tissue from individual patients, the assay was able to extrapolate a pattern of gene activity levels, yielding individual risk scores that were converted to one of three risk categories (high risk, intermediate risk or low risk) based on predetermined assay cutoff values.

After the completion of Kaplan-Meier survival analysis, the Kaiser Permanente validation cohort demonstrated 5-year survival of 71.4% (95% CI, 60.5-80) in the low-risk group, 58.3% (95% CI, 48.9-66.6) in the intermediate-risk group and 49.2% (95% CI, 42.2-55.8) in the high-risk group. A sensitivity analysis — excluding 18 patients in the cohort who received adjuvant chemotherapy — provided similar 5-year survival outcomes: 70% (95% CI, 58.7-78.8) in the low-risk group, 58.2% (95% CI, 48.5-66.7) in the intermediate-risk group and 48.9% (95% CI, 41.7-55.6) in the high-risk group.

Comparable analysis from the CCTC cohort exhibited 5-year OS of 74.1% (95% CI, 66-80.6) in low-risk, 57.4% (95% CI, 48.3-65.5) in intermediate-risk and 44.6% (95% CI, 40.2-48.9) in high-risk patients. According to researchers, the assay improved predictive accuracy beyond National Comprehensive Cancer Network criteria for stage I high-risk tumors (P,.0001) and was able to distinguish low-risk, intermediate-risk and high-risk patients within all disease stages.

“The CCTC represents a revolutionary new collaborative approach to clinical research among Chinese and Western experts,” researcher Michael J. Mann, MD, a UCSF associate professor of cardiothoracic surgery, said in a press release. “It is fitting that this body’s first major effort may support a molecular personalization of lung cancer care, and yield one of the first examples that fundamental tumor biology reaches across ethnic lines and can be used to try to improve outcomes for a large percentage of patients.”

Disclosure: The researchers report consulting relationships with Pinpoint Genomics Inc.

Nathan Pennell, MD, PhD
Nathan
Pennell

Although we can improve the cure rate of resected stage II and III non–small cell lung cancer patients with adjuvant chemotherapy, patients with stage I disease remain at risk for recurrence but have no data to support benefit from adjuvant treatment. The study by Dr. Kratz and colleagues describes a new and well-validated prognostic assay to distinguish patients with stage I nonsquamous NSCLC who are at high risk of recurrence and death from those at intermediate or low risk. Although there are a number of prior studies that have developed similar assays (Subramanian J. J Natl Cancer Inst. 2010;102:464-474), the current study is distinct in that it can be performed in widely available paraffin-embedded tissue and in the extent of external validation, indicating that the results are quite reliable.

The major issues with this assay are the exclusion of patients with squamous cell carcinoma (30% of all NSCLC cases), and that this is a prognostic signature rather than one predictive of benefit from adjuvant chemotherapy. The authors acknowledge this issue and plan a prospective study to test their assay in resected stage I patients, although this will likely take nearly 10 years to complete. To date, there has been only one prognostic assay in early-stage NSCLC that was also shown to be predictive of benefit from chemotherapy in the high-risk group (Zhu CQ. J Clin Oncol. 2010;28:4417-4424), but interestingly, there is significant overlap in the genes used in both signatures, suggesting that the new assay may well end up being a useful test to identify patients who would benefit from adjuvant chemo who would otherwise not receive it under current treatment standards. Until this has been prospectively validated, however, assays such as this one remain experimental and should not be used to guide therapy.

Nathan Pennell, MD, PhD
Medical Oncology, Internal Medicine
Taussig Cancer Center
Cleveland Clinic

Follow HemOncToday.com on Twitter.