Smoking is a risk factor for cervix cancer and causes hypoxemia, which promotes tumor infiltration and potentially impacts on treatment outcome. We performed a retrospective study to determine if smokers had an increased risk of uterine corpus infiltration, which is associated with more advanced disease and/or treatment failure after primary chemoradiation.
Results from a prospective database of patients treated with primary chemoradiation for locally advanced cervix cancer with a pretreatment MRI were analyzed. Smoking status was assessed by self-report at presentation.
Smoking status was recorded for 346 of the 362 patients with 98 current smokers (28%), 56 ex-smokers (16%), and 192 nonsmokers (55%). Median age was 58 years with ever-smokers having a younger age at diagnosis than nonsmokers. Histologic type, International Federation of Gynecology and Obstetrics stage, tumor volume, and nodal involvement were similar across groups, as were toxicities of treatment. Ever-smokers were more likely to have corpus uterine invasion than nonsmokers. Ever-smokers had more recurrences than nonsmokers, with nonsmokers having a longer median overall survival (50.1 vs 38.7 months, P = 0.004) and relapse-free survival (46.8 vs 28.5 months, P = 0.003). In multifactor analysis, ever-smoking status was a significant predictor of developing corpus invasive disease and of inferior relapse-free and overall survival after treatment.
Smokers have a greater risk for developing corpus invasive cervix cancer. Although nonsmokers have an older age at diagnosis, they live longer and have fewer recurrences after a diagnosis of locally advanced carcinoma of the cervix.
*Peter MacCallum Cancer Centre, Victoria, Australia; †The University of Melbourne, Victoria, Australia; and ‡Rural Clinical School, The University of Queensland, Australia.
Address correspondence and reprint requests to Linda Mileshkin, MD, FRACP, Peter MacCallum Cancer Centre, St Andrews Place East, Melbourne, Victoria, Australia 3002. E-mail: Linda.Mileshkin@petermac.org.
The authors declare no conflicts of interest.
Received January 9, 2014
Received in revised form April 1, 2014
Accepted April 9, 2014