Presenter: S.I.Ou Presenter's Affiliation: Chao Family Comprehensive Cancer Center, University of California, Orange, CA; University of California Irvine, Irvine, CA Type of Session: Scientific
Non-small cell lung cancer (NSCLC) remains the number one cause of cancer death in both men and women in the United States. The greatly increased risk of NSCLC development in tobacco users has been well recognized for several decades.
In 2002, Nordquist and colleagues from the H. Lee Moffit Cancer Center in Tampa, Florida, presented findings of improved survival in never-smokers with NSCLC versus current smokers with NSCLC at the American Society of Clinical Oncology Annual Meeting.
Since that time, NSCLC in never-smokers has come to be recognized as an entity separate from NSCLC in former and current smokers.
A never-smoker is defined in the literature as an individual who has smoked less than 100 cigarettes in his or her lifetime.
This is in contrast to a former smoker, an individual who has smoked at least 100 cigarettes, but stopped smoking over one year ago, or a current smoker, who continues to smoke.
NSCLC in never-smokers now represents the 13th most common cancer world wide (surpassing primary brain cancer and ovarian cancer in incidence), and is the 7th leading cause of cancer death globally.
The biology of NSCLC in never-smokers is recognized as different from that of NSCLC in smokers (Marks, 2008).
Mutations affecting the epidermal growth factor receptor (EGFR) and the Ras oncogene are known to contribute to development of NSCLC tumors.
Mutations in these genes are present in the tumors of 31% of current and former smokers with NSCLC. In this cohort, Ras mutations are more frequent than EGFR mutations.
Mutations are present in a much greater percentage of tumors that arise in non-smokers (61%). Additionally, the proportions of mutations are reversed in this group, with EGFR mutations being much more common than Ras mutations.
Recently, common features of never-smokers with NSCLC have been identified, including East Asian ethnicity, female sex, adenocarcinoma (specifically bronchoalveolar) pathology, and increased likelihood of EGFR mutations (Ho, 2005).
Large population-based studies have been limited in the United States, however, largely because the Surveillance Epidemiology and End Results (SEER)/ Medicare databases do not include information on smoking status. Because these databases are frequently used for population-based analysis of the United States population, information on smoking status has been frequently excluded from these studies.
The study presented here was performed with the intent of increasing understanding of the epidemiology of NSCLC in never-smokers, and to elucidate the impact of ethnicity on prognosis for patients with this disease.
Materials and Methods
The Cancer Surveillance Program’s patient database from three Southern California counties (Orange County, San Diego County, and Imperial County) was used as the data source for this retrospective study.
A data gathering algorithm was used to generate the patient cohort examined in this study from the larger database.
25,204 NSCLC patients diagnosed with NSCLC between 1991 and 2005 with known smoking status were identified, and served as the population cohort examined.
Prognostic factors examined included age, gender, histology, histologic grade, and ethnicity, which had been reported to the Cancer Surveillance Program by patients’ medical providers.
Other information gathered included cancer stage and modalities employed for treatment.
Statistical analyses of outcomes between never- and former/ current smokers were performed using Chi-square and the Kaplan-Meier method.
Of 25,204 patients analyzed, 2240 (8.9%) were never-smokers.
Median age at lung cancer diagnosis was 71 years in never-smokers, and 69 years in smokers (p < 0.001).
In subset analysis, disease onset occurred at older median age across ethnicities, with the exception of Hispanics: In this group, median age of onset was lower in never-smokers than smokers (67 years versus 68.5 years, respectively, p < 0.001).
Overall survival and lung cancer-specific survival were significantly improved in never-smokers versus smokers (p < 0.0001, p < 0.0001, respectively), regardless of gender.
On multivariate analysis, smoking status was a poor prognostic factor, with hazard ratio of 1.14 for overall survival and 1.16 for lung cancer-specific survival. Overall survival benefit appeared to be independent of cancer stage.
Never-smokers appeared to have improved treatment response, with improved overall survival in never-smokers undergoing surgical treatment versus smokers. Never-smokers with metastatic disease also appeared to respond better to chemotherapy than smokers, with a 27% absolute improvement in overall survival, and 8 month median survival in never-smokers versus 6 month median survival in smokers with metastatic disease receiving chemotherapy. Although data regarding response to radiation were not formally presented, the authors noted that never-smokers with locally advanced NSCLC undergoing definitive chemoradiation also appeared to have improved outcomes.
When the two groups were compared (never-smokers versus smokers), the group of never-smokers with NSCLC contained a higher proportion of women. Tumors in never smokers were also more likely to be adenocarcinomas, and were three-times as likely to be bronchoalveolar carcinomas.
Of Caucasian women with NSCLC, approximately 13% were never-smokers. This is in contrast to 45% of Asian women with NSCLC and 25% of Hispanic women with NSCLC. Of African-American women with NSCLC, 11% were never-smokers.
Of women born in North America, 11% of women born in Canada and 12% of women born in the United States with NSCLC were never-smokers. Thirty percent of women born in Mexico with NSCLC were never smokers.
Of women born in Asian countries with NSCLC, 53% of Philippine women, 64% of Vietnamese women, 17% of Japanese women, and 61% of Chinese women were never-smokers.
Asian ethnicity appeared to be a favorable prognostic sign in both smokers and never-smokers. For overall survival, hazard ratios were 0.847 for smoking Asians versus smoking non-Asians, and 0.861 for never-smoking Asians versus never-smoking non-Asians.
The authors conclude that never-smoking status and Asian ethnicity are favorable prognostic factors in patients with NSCLC, the latter regardless of smoking status.
They note that among female NSCLC patients, the group of Asian women presented here contained the highest percentage of non-smokers, followed in decreasing order by Hispanic, Caucasian, and African-American women.
This study represents an interesting epidemiological survey of NSCLC patients living in California, focusing on issues of ethnicity and smoking status.
The results of this study confirm findings documented by other groups regarding increased proportions of women and Asians among never-smokers diagnosed with NSCLC.
The proportion of never-smoking Asian women developing NSCLC appears to be increased in women born in Asian countries compared to those born in North America. Further investigation of this issue would be of great interest as we continue to explore issues of exposure versus genetics in the role of cancer development. The majority of NSCLC patients have a known exposure to cigarette smoke and related carcinogens; the never-smoking population represents a unique subset without this known exposure. Increased understanding of differences in cancers of never-smokers based on birthplace could be potentially beneficial in furthering our understanding of events promoting cancer development.
The increased proportion of never-smokers among Hispanics diagnosed with NSCLC identified in this study is a new piece of information with implications that are potentially very important.
The biology of NSCLC tumors that develop in never-smokers has been examined by other groups; however, specific analysis of potential biologic differences among never-smoking Hispanics with NSCLC versus Asians would certainly be of interest.
In this study, never-smokers appeared to have increased response to major treatment modalities, including surgery, radiotherapy, and chemotherapy. Additionally, other groups have demonstrated that never-smokers are more likely to demonstrate improved response to the biologic agents gefitinib (Iressa) and erlotinib (Tarceva), selective inhibitors of the EGFR tyrosine kinase domain. This improved response in all likelihood contributes to improved prognosis; however, it should be noted that the details of cancer treatment were largely not addressed in this study. Patients with NSCLC represent a heterogenous group, and treatment of patients with locally-advanced NSCLC was not formally addressed within this study. Certainly, a more detailed analysis of response to radiotherapy and specific chemotherapy agents would be interesting and important to future tailoring of treatment regimens.
This study is somewhat limited in terms of conclusions regarding prognostic factors due to its retrospective and population-based nature. The epidemiologic information presented is, however, interesting in and of itself. The proportion of non-smokers among women diagnosed with NSCLC in certain Asian subtypes is quite high, and further understanding of the biology leading to the development of this disease, as well as available and effective treatments, is certainly valuable in both treatment of this subtype of patients and improving our understanding of NSCLC as a whole.