Abstract 592: The potential effect of smoking cessation on mortality among patients diagnosed with smoking-related cancers

Abstract

Abstract Background: The benefit of smoking cessation on mortality after cancer diagnosis is unclear, particularly given ethical restrictions of randomizing patients who are smokers to a smoking cessation intervention or continued smoking. We thus aimed to assess the effect of a hypothetical smoking cessation intervention on mortality among patients diagnosed with smoking-related cancers using observational data. Methods: We used data from the JPS Oncology Registry, for which the source population is cancer patients in Tarrant County, TX who received care at a public health care system regardless of their ability to pay. Our eligible population included patients diagnosed with primary lung, head and neck, or colorectal cancer between January 2008 and June 2013 with follow-up through 2016, and self-reported current or former use of cigarette, cigar, or pipe tobacco products. We used a potential outcomes framework to estimate cancer-specific effects of a hypothetical smoking cessation intervention on 1- and 3-year all-cause mortality. Under certain assumptions, the estimated marginal risk ratios and risk differences from this approach are analogous to intention to treat estimates from a randomized controlled trial. Results: Our study population comprised 1,058 patients diagnosed with lung (n=582), head and neck (n=248), or colorectal cancer (n=228). The median age at diagnosis overall was 57 years (interquartile range=51-63). Patients were predominantly male (64%), non-Hispanic White (63%), and uninsured (50%). Modeled smoking cessation decreased mortality for head and neck cancers but not lung or colorectal cancers (Table 1). Conclusion: Albeit with certain limitations, our findings suggest that a hypothetical smoking cessation intervention initiated at the time of cancer diagnosis may not have homogeneous effects on mortality across cancer types. These results provide insight about target populations for whom smoking cessation may benefit cancer survival. Table 1. Risk ratios and differences for the effect of hypothetical smoking cessation intervention1-year mortality3-year mortalitySiteRRa (95% CL)bRDc (95% CL)RR (95% CL)RD (95% CL)Lung0.98 (0.86, 1.1)-0.01 (-0.10, 0.08)1.0 (0.95, 1.1)0.02 (-0.05, 0.08)Head and Neck0.71 (0.45, 1.1)-0.11 (-0.25, 0.03)0.78 (0.58, 1.0)-0.14 (-0.29, 0.01)Colorectal0.87 (0.48, 1.6)-0.03 (-0.16, 0.10)1.1 (0.73, 1.5)0.02 (-0.14, 0.19)aRisk Ratio < 1 indicate lower risk of mortality if all patients adhered to a hypothetical smoking cessation intervention.bConfidence LimitscRisk difference < 0 indicate a lower risk of mortality if all patients adhered to a hypothetical smoking cessation intervention. Citation Format: Bradford E. Jackson, Tracey E. Barnett, Yan Lu, Bassam Ghabach, Rohit P. Ojha. The potential effect of smoking cessation on mortality among patients diagnosed with smoking-related cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 592.