Studies Determine Smokers' Risks of Recurrence, Progression, and Survival in Bladder Cancer

Shahrokh F. Shariat, MD

An international group of investigators provided first-time evidence of the impact of smoking on outcomes in bladder cancer in results from two studies that were reported at the 2012 Society for Urologic Oncology) annual meeting in Bethesda, Maryland. Both studies found significant increased risks of recurrence, disease progression, and survival with continued smoking after diagnosis and treatment for bladder cancer. The studies also quantified a timeframe for reduction of these risks to that of nonsmokers following smoking cessation.

Smoking and Smoking Cessation in Primary Nonmuscle-Invasive Bladder Cancer (NMIBC)

Smoking has long been a known risk factor for bladder cancer, but a dose-response relationship has not previously been well described in the literature.Smoking status and higher cumulative smoking exposure were linked to worse prognoses in 2043 patients with NMIBC who were treated with transurethral resection of the bladder with or without intravesical instillation therapy, as reported by Evanguelos Xylinas, MD, from Weill Cornell Medical College in New York City.

Smoking histories that were obtained from patients included an average number of cigarettes smoked per day (CPD); duration of smoking in years; and time since smoking cessation. Lifetime cumulative smoking exposure was designated in the following manner: light short-term as ≤19 CPD for ≤19.9 years; light long-term as ≤19 CPD for ≥20 years. Heavy short-term smoking exposure was defined as ≥20 CPD for ≤19 years; and heavy long-term smoking exposure was defined as ≥20 CPD for ≥20 years.

The results showed that cumulative smoking exposure among both current and former smokers was significantly associated with disease recurrence (P <.001), disease progression (P <.001), and overall survival (P <.001) in multivariate analysis. Heavy long-term smokers had the worst outcomes.

However, smoking cessation of greater than 10 years’ duration significantly reduced the risk of disease recurrence (hazard ratio [HR] = 0.66; 95% CI, 0.52-0.84; P <.001) and progression (HR = 0.42; 95% CI, 0.22-0.83; P = .036) in multivariate analysis (Table).

Another similar study, examining the impact of smoking and smoking cessation on outcomes in bladder cancer patients who were treated with radical cystectomy, produced some similar results.

Table. Smoking-Related Outcomes in Primary Nonmuscle-Invasive Bladder Cancer

Disease Recurrence

Disease Progression

HR (95% CI)

P value

HR (95% CI)

P value

Age (years)

1.00 (0.99, 1.01)

0.760

1.00 (0.98, 1.02)

0.980

Gender

0.920

0.820

Male

1.00

1.00

Female

0.99 (0.82, 1.20)

1.05 (0.69, 1.6)

Smoking status

(including

cessation status)

.001

0.036

Current

1.00

Former < 10 years

1.30 (1.09, 1.53)

0.99 (0.65, 1.50)

Former ≥ 10 years

0.66 (0.52, 0.84)

0.42 (0.22, 0.83)

Cumulative

smoking exposure

<.001

<.001

Heavy-long

1.00

1.00

Light-long

0.91 (0.77, 1.07)

0.43 (0.29, 0.63)

Heavy-short

0.43 (0.30, 0.60)

0.12 (0.03, 0.44)

CI indicated confidence interval; HR, hazard ratio.

Impact of Smoking and Smoking Cessation on Outcomes in Bladder Cancer Patients Treated With Radical Cystectomy

Smoking was associated with worse prognosis among 1506 patients who were treated with radical cystectomy (RC) for urothelial carcinoma of the bladder (UCB) in results that were dose-dependent. Lifetime cumulative smoking exposure was categorized in the same manner as the companion study in NMIBC. (All patients underwent RC without receiving preoperative chemotherapy.)

Smoking status was associated with a cumulative incidence of disease recurrence (P = .004) and cancer-specific mortality (P = .016) in univariate analyses, and disease recurrence in multivariate analysis (P = .02). Among patients who had ever smoked, cumulative smoking exposure was associated with advanced tumor stages (P <.001), lymph node metastasis (P = .002), disease recurrence (P <.001), cancer-specific mortality (P = .001), and overall mortality (P = .04). Heavy long-term smokers experienced the worst outcomes. These effects were significantly mitigated by smoking cessation longer than 10 years’ duration.

These two studies are part of a series of five that examine the effects of smoking on disease outcomes, according to Shahrokh F. Shariat, MD, assistant professor of Urology at the Weill Cornell Medical Center, who is providing leadership for the series.

Smokers are not only at higher risk for higher-stage bladder cancer, but they are also less likely to respond to bacillus Calmette Guérin (BCG) immunotherapy for high-risk NMIBC, said Shariat, commenting on the studies and their implications. Smokers also have a higher risk of disease progression to muscle-invasive disease, requiring removal of the bladder. And, persons who continue to smoke after bladder removal have a reduced chance of achieving an outcome that results in a cure. Instead, he said, for these patients, “…disease recurrence is higher, rates of progression are higher, and there is a greater likelihood for the development of metastasis and death from cancer.”

Their efforts to better understand why current smokers fared worse than past smokers and to identify which ones did well resulted in a finding similar to the cardiovascular literature. “We found that if you stop smoking for over 10 years [prior to a diagnosis], your risk reverts to that of never-smokers.” People who stop smoking for one year or two years may not benefit at all, however, because it seems that the molecular alterations that resulted from the smoking continue to exert an impact for an unknown length of time. But at 10 years, “…this risk appears to revert back to that of never-smokers.”

“

We found that, if you look at the different cut-offs, smoking more than 20 cigarettes per day for about 20 years seems to be an important threshold”

—Shahrokh F. Shariat, MD

In designing these studies, the research group was fortunate to have a homogeneous dataset with different disease stages and large series. But they rejected the “pack-years” scheme to arrive at a dose-exposure risk that distinguished between persons who smoke heavily versus lightly and those who smoke one or two cigarettes a day. Pack-year, which combines the cigarettes smoked per day and the duration of smoking, assumes that the “dose” of smoking one cigarette per day for 10 years is the same as the dose of 10 cigarettes per day for one year. With their better, more predictive formula, Shariat said, “We found that, if you look at the different cut-offs, smoking more than 20 cigarettes per day for about 20 years seems to be an important threshold.” People who smoke more than 20 cigarettes per day for more than 20 years constitute the worse group of all. There is not only a relationship between smoking and outcomes, but there seems to be a dose-response: the higher the length of exposure and the higher the dose of exposure, the worse the prognosis.

This work is the first to show that smoking not only affects the development of bladder cancer—the process of carcinogenesis—but also progression of the disease, indicating that there is activation of certain molecular pathways. Shariat and colleagues are now conducting molecular translational studies to look at markers such as polymorphisms and DNA repair genes, aiming to extend an understanding of smoking effects at the most basic level.

For patients, what has resulted from this research is that now— thanks to electronic medical records— smokers are referred to smoking cessation programs. Patients who say “I already have cancer, why should I stop now” can be told, with the strength of scientific evidence, that if they continue to smoke they are going to do worse.