Lung cancer remains the leading cause of cancer mortality, accounting for over 160,000 deaths in the United States and 1.18 million deaths worldwide each year.[1,2] Though tobacco smoking remains the most strongly associated risk factor for the development of lung cancer, 10% to 15% of lung cancer patients in the United States lack any history of tobacco use.[1,3,4] The development of lung cancer in nonsmokers is of even greater concern worldwide, with some estimates suggesting that as many as 15% of cases in men and 53% of cases in women may be unrelated to tobacco smoking. In this issue of ONCOLOGY, Subramanian and Govindan address this important topic, discussing the epidemiology, molecular biology, and clinical outcomes of lung cancer in nonsmoking patients.
While it is unclear whether prevalence of lung cancer among nonsmokers is growing, even at its present levels there are an estimated 16,000 to 24,000 deaths from lung cancer in nonsmokers each year in the United States. Regarded as its own entity, lung cancer in nonsmokers would itself be among the top 10 causes of cancer mortality in this country. It will be critical for scientists to explore the epidemiology of lung cancer in nonsmokers in order to learn more about the potential genetic and environmental risk factors that underlie this phenomenon.
A World Apart
There are important differences between smokers and nonsmokers in the epidemiology and clinical course of lung cancer. Retrospective studies completed prior to the advent of widespread use of oral tyrosine kinase inhibitors (TKIs) of the epidermal growth factor receptor (EGFR) demonstrate that lung cancer in “never-smokers” is more closely associated with adenocarcinoma histology, female gender, and an earlier age of onset.[6,7] Moreover, nonsmokers with lung cancer may have improved outcomes compared with smokers. In resected early-stage disease, nonsmokers had improved postoperative survival as well as superior 10-year overall and disease-specific survival.[8,9] In metastatic disease, the impact of smoking on clinical outcomes to conventional chemotherapy is less clear, with some studies suggesting a benefit in favor of never-smokers, while others show no significant difference in outcomes based on smoking status.
Perhaps the most important difference between lung cancers in nonsmokers and smokers has been in the incidence of specific genomic changes. Subramanian and Govindan discuss the increased likelihood for nonsmokers to harbor an EGFR mutation or the EML4-ALK translocation. In addition, not only are tumors from smokers more likely to harbor a KRAS mutation, but the types of KRAS mutations found in smokers vs nonsmokers varies: smokers are more likely to harbor a transversion mutation in that gene (ie, substituting a purine for a pyrimidine, or pyrimidine for purine), while KRAS mutations in nonsmokers are more likely to be transition mutations (ie, a substitution of one purine to another, or one pyrimidine to another). The importance of these genomic changes not only speaks to a difference in the underlying biology of tumors in nonsmokers vs smokers, but it also points us directly toward critical differences in treatment strategies and outcomes based on tumor genome.
Moving Beyond Clinical Characteristics
The Iressa Pan-Asia Study (IPASS) and other recent publications provide critical insight into the impact of clinical and molecular markers in determining the most appropriate first-line treatment strategy for each patient. In the IPASS trial, Asian patients with advanced pulmonary adenocarcinoma who had smoked a total of less than 10 pack-years were randomized to receive either the EGFR-TKI gefitinib(Drug information on gefitinib) (Iressa) or conventional chemotherapy with carboplatin(Drug information on carboplatin) plus paclitaxel(Drug information on paclitaxel). While the overall results showed a similar median progression-free survival (PFS) for the gefitinib group (5.7 months) and carboplatin/paclitaxel group (5.8 months), a planned subset analysis by EGFR mutation status provided a critical observation: PFS was significantly improved with gefitinib rather than carboplatin/paclitaxel (hazard ratio [HR] = 0.48; 95% confidence interval [CI] = 0.36–0.64; P < .001) in patients with an EGFR mutation, while PFS was significantly worse with gefitinib compared to conventional chemotherapy in patients with wild-type EGFR (HR = 2.85; 95% CI = 2.05–3.98, P < .001).
These IPASS results suggest that even in a clinically enriched population (Asian ethnicity, adenocarcinoma histology, tobacco history < 10 pack-years), EGFR mutation status is a much better predictor of who will benefit from therapy with an EGFR-TKI. Among 91 patients in the IPASS trial who were treated with gefitinib with known wild-type EGFR status, there was only one objective response to therapy. This position is further supported by results of an individual patient meta-analysis of a database incorporating five clinical trials of first-line EGFR-TKI therapy. In that study, a comparison of clinical to molecular characteristics demonstrated superiority for EGFR mutation status over any collection of clinical characteristics in identifying a group of patients most likely to benefit from therapy with an EGFR-TKI.
With the IPASS findings and other results, it is more difficult to justify first-line therapy with an EGFR-TKI based on clinical characteristics alone. In doing so, we would inappropriately direct never-smokers who are EGFR wild-type toward an EGFR-TKI when they might be better served by conventional chemotherapy. Similarly, we might dismiss a smoker with adenocarcinoma, when that patient may in fact harbor an EGFR mutation that would otherwise be associated with sensitivity to gefitinib or erlotinib (Tarceva) therapy. Therefore, we need to move beyond treatment based on clinical characteristics alone and transition toward more routine genetic testing for our newly diagnosed patients. Because changes such as EGFR mutations and EML4-ALK translocations are seen almost exclusively in adenocarcinomas, it is reasonable to narrow routine testing in that fashion.
Changing Clinical Practice
Adopting routine genetic testing into our clinical practice will require several changes. First, we need to continue to refine our methods of genetic testing, in order to reduce both the amount of tissue needed and the time required to report results. Until this is achieved, though, we need to move toward initial diagnostic procedures that can also provide sufficient tissue for the current commercially available methods of genetic testing (ie, core needle biopsies rather than fine-needle aspirates) whenever this can be done safely. Obtaining sufficient tissue at diagnosis obviates the need for subsequent procedures and allows for earlier genetic testing, so that results can be obtained in a more timely fashion and can inform first-line therapy choices. With genetic information in hand, clinicians can then incorporate results from studies like the IPASS trial into everyday practice, personalizing treatment choice based on specific genetic changes in an individual’s tumor.
Financial Disclosure: The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.