Abstract Body

Cancers are a major source of morbidity and mortality in the cART era. The prevalence of smoking in HIV+ people is 40–70% and the clinical benefits of smoking cessation on cancer risk have not been reported. We aimed to estimate cancer rates after smoking cessation in persons from the D:A:D study.

Persons were followed from the latest of study entry or 1/1/2004 until earliest of first cancer diagnosis, last visit plus 6 months, death, or 1/2/2015. Three outcomes were considered: all cancers combined, lung cancer, and other smoking-related excluding lung cancer (OSRC; see footnote). Smoking status was defined as current and never smokers, those who stopped during follow-up (<1,1-2,2-3,3-5,5> years since stopping) and those who stopped prior to baseline. Adjusted rate ratios (aRR) were calculated using Poisson regression.

39701 persons contributed 315327 person years of follow-up (PYFU) (median: 9 IQR: 6, 11 years per person). At baseline, 41% of people were current smokers, 17% were ex-smokers, 27% never smoked. 2230 developed cancer (IR 7.1/1000 PYFU, 95%CI: 6.8, 7.4), of which 251 were lung cancers (IR 0.8/1000 PYFU, 95%CI: 0.7, 0.9) and 516 were OSRC (IR 1.6/1000 PYFU 95%CI: 1.5, 1.8). Incidence of all cancers combined (Figure) was highest <1 year after quitting compared to those who had never smoked (aRR: 1.62 95%CI: 1.32, 1.99) and was similar to never smokers thereafter. Lung cancer incidence was over 11-fold higher <1 year after quitting (aRR: 11.72 95%CI: 4.81, 28.57) and remained >8-fold higher even after 5 years (aRR: 8.26 95%CI: 2.83, 24.09) with no evidence of decline when compared to non-smokers. OSRC incidence was almost 3-fold higher <1 year after quitting (aRR2.52 95%CI: 1.69, 3.74), but was similar to never smokers thereafter. Smoking duration was associated with the occurrence of all cancers combined (Per year longer aRR: 1.03 95%CI: 1.01,1.04), lung (aRR: 1.07 95%CI: 1.01, 1.12), but not OSRC (aRR: 1.03 95%CI: 0.99, 1.06). No significant interactions between smoking status and age, gender or CD4 were found.

Overall cancer incidence declined to that of non-smokers after one year quitting except for lung cancer incidence, which did not decrease even >5 years after quitting. Smoking cessation efforts should be a priority to reduce the risk of cancer, however, surveillance and screening of lung cancer should not be stopped in patients who stop smoking.