Smoking Worsens Prognosis in Cervical Cancer Patients

March 1, 2003

NEW ORLEANS-In women with locally advanced cervical cancer, smoking is an independent risk factor for a significantly worse outcome, a new study from the Gynecologic Oncology Group (GOG) has shown.

NEW ORLEANS—In women with locally advanced cervical cancer, smoking is an independent risk factor for a significantly worse outcome, a new study from the Gynecologic Oncology Group (GOG) has shown.

Smoking has previously been shown to be an independent risk factor for squamous cell carcinoma of the cervix. About 50% of North American women who develop cervical cancer, in fact, are smokers. The present study takes this finding a step further, showing that smoking also affects survival.

GOG 165

In GOG 165, reported at the 34th Annual Meeting of the Society for Gynecologic Oncologists (abstract 71), 315 patients with stage IIb, IIIb, and IVa cervical cancer were randomized to two different treatment arms from 1977 to 2000 and followed for a median of 31.5 months. GOG 165 compared the effectiveness of cisplatin (Platinol)-based chemoradia-tion vs chemoradiation using prolonged venous infusion of fluorouracil (5-FU). The duration of radiotherapy was similar for the two groups, and gastrointestinal toxicity was also similar.

Prior to beginning chemoradiation therapy, patients completed a questionnaire pertaining to past and current smoking behavior. Subjects were considered smokers if they were currently smoking one or more cigarettes per week. Smoking status was confirmed by urinary cotinine analysis, with a level of 100 ng/mL or greater being considered indicative of smoking. Self-reported smoking and urinary cotinine results had 94.4% agreement. Of the 315 patients, 42% were current smokers. Current smoking status was associated with younger age, higher stage of disease, and squamous histology.

Study Results

The study found a 44% increased risk of disease progression and a 59% increase in the relative risk of death among current smokers, reported Steven E. Waggoner, MD, of Case Western Reserve University, Cleveland.

In smokers, by unadjusted analysis, the hazard ratio was 1.44 for progression-free-survival (P = .038) and 1.59 for overall survival (P = .021). In the adjusted analysis, the hazard ratio was 1.33 for progression-free survival (P = .069) and 1.69 for overall survival (P = .022), Dr. Waggoner said.

"Smoking was associated with a worse overall survival in these patients with cervical cancer. This cannot be explained by differences in other risk factors," Dr. Waggoner said. "There were more smokers with stage III disease, but when we controlled for stage, we still found smoking to be an independent risk factor."

The biologic mechanisms responsible for this association are unknown, but may involve a tobacco-mediated molecular pathway and include factors such as tumor hypoxia that might be positively influenced by smoking cessation, Dr. Waggoner said.

"It’s important to note that 88% of the deaths in this study so far are directly attributable to cancer or cancer treatment. The 12% of cases that are due to other causes are evenly divided between smokers and nonsmokers, so there does not appear to be greater mortality from noncancer causes in the smoking group," he added.

Discussant Carolyn Muller, MD, of The University of Texas Southwestern Medical Center, Dallas, said that one of the study’s strengths was the requirement of a biomarker correlate to support the self-reported smoking data. However, she noted that the clinical and biochemical thresholds for defining "smoker" were somewhat arbitrary, and the requirement of only two urinary cotinine levels "may not adequately represent tobacco exposure throughout the treatment."

Of interest was the finding that smoking significantly affected overall survival in the multivariate analysis but not progression-free survival (where only a trend was demonstrated)—a finding that runs counter to most treatment studies performed by the GOG, she noted.

"Can this observation be explained by determining the mechanism by which tobacco exposure alters the tumor or the host?" she asked. "For instance, one could speculate that a more aggressive carcinogenic pathway may take place as cervical cancer develops in chronic smokers."

This theory is supported by work from Dr. Muller’s group, who reported results at SGO (abstract 35). Their study showed that aberrant p16 methylation was strongly associated with active tobacco use in squamous cell cervical cancers. Previous studies have shown that the p16 gene, which is located on chromosome 9, plays a key role in controlling cell growth, and its aberrant methylation is strongly associated with tobacco exposure.

In the comparison of cervical cancer vs normal cervical cytological specimens, Dr. Muller and her colleagues showed that aberrant p16 methylation was significantly higher in cervical cancers (62%) than in normal specimens (10%). Smokers in her study had a 20-fold risk of having aberrant p16 methylation, compared with nonsmokers, she said.