Chapter 4: Cancer incidence in Canada: trends and projections (1983-2032) – Lung cancer - HPCDP: Volume 35, Supplement 1, Spring 2015

Chapter 4: Projections by Cancer Site

8. Lung cancer

Lung cancer is the second most common cancer and the leading cause of cancer death in both males and females in Canada. One in 11 males and 1 in 15 females can expect to be diagnosed with lung cancer in their lifetime, and 1 in 13 males and 1 in 18 females can expect to die from it.^{Endnote 1} The average annual number of new cases in 2003–2007 was 12 245 in males and 9865 in females (Tables 4.8.1 and 4.8.2), accounting for 15.2% and 13.3% of all male and female cancer cases, respectively (Figure 3.9). Lung cancer has a poor 5-year relative survival rate in Canada, at 14% in males and 20% in females for 2006–2008.^{Endnote 1}

In 2003–2007, the incidence rates increased with age to peak in the 75–84 age group, at 567.3 per 100 000 in males and 297.6 per 100 000 in females (Tables 4.8.3 and 4.8.4).

The overall lung cancer rates were lower in males under age 55 than in females for 2003–2007, and the rates in males aged 55 or older were higher than the female rates in the whole observation period (Figure 4.8.2). The male-to-female ratio of incidence rates increased with age to 2.5:1 in people aged 85 or older in 2003–2007.

Lung cancer incidence rates for males have dropped over the entire observation period and showed a slowing increase for females (Figure 4.8.1). Between 1998 and 2007, the lung cancer incidence rates decreased significantly in males by 1.8% per year but increased significantly in females by 1.1% per year (Figures 3.1 and 3.2). All the regions have had a consistent decrease over the whole observed period for males. Rates for females have increased in each region although they have stabilized in British Columbia since 1993–1997 and in Ontario 5 years later (Figure 4.8.1).

The analysis predicts a steeper decrease in rates for males in the east (Ontario, Quebec and the Atlantic region) than in the west (Figure 4.8.1). For females, the rates will decrease in Ontario and British Columbia, and increase until 2013–2017 in Quebec and until 2008–2012 in other regions, followed by a gradual decrease. This projected future downturn of the rates is because of a decrease in rates in later birth cohorts. Quebec is predicted to continue to have the highest rates of lung cancer in both sexes. The differences in ASIRs in females between Quebec and the other regions will increase from between 10% and 30% in 2003–2007 to between 33% and 65% in 2028–2032. These increased differences are not evident in males.

The male-to-female ratio of ASIRs for lung cancer in Canada has decreased from 3.1:1 to 1.5:1 over the observed period (1983–2007), and is projected to further decrease to 1.2:1 in 2028–2032 (Figures 4.8.1 and 4.8.2).

From 2003–2007 to 2028–2032, the ASIRs of lung cancer for Canada are projected to decrease in males by 34%, from 70.7 to 46.4 per 100 000, and to peak and then also decrease in females by 16%, from 47.1 to 39.6 per 100 000. Because of the aging and growth of the population, the annual number of new cases is projected to increase by 34% in males (from 12 245 to 16 420) and by 62% in females (from 9865 to 15 945).

Comments

Smoking causes about 90% of lung cancer deaths in males and between 75% and 80% in females in the US.^{Endnote 128, Endnote 129} The trends in lung cancer incidence rates in Canada have closely mirrored historical patterns of smoking prevalence,^{Endnote 42, Endnote 43} after accounting for a latency period of 20 years or more. Because smoking prevalence began to decrease in the mid-1960s in males and in the mid-1980s in females,^{Endnote 42, Endnote 43} lung cancer incidence rates decreased for males over the whole observation period and showed the slowing of increase for females. The rates of lung cancer decreased from 1993–1997 to 2003–2007 in females aged less than 45, were relatively stable in women aged 45 to 54, showed a slowing increase in those aged between 55 and 64, and increased in those aged 65 or older. These trends are in agreement with the fact that the onset of cigarette smoking starts at younger ages and then follows the birth cohort as it ages, and lung cancer rates increase as the birth cohort ages.

Figure 4.8.3 shows the different results for projected lung cancer incidence in males and the similar results for females based on age–period–cohort models that include and exclude adjustment for available smoking prevalence data in Canada. In males, for the models with the same parameter settings (cut trend and using recent 10-year slope), the models incorporating smoking rates predict lower lung cancer incidence rates as compared with the models without adjustment for smoking, echoing the decreased pattern of tobacco consumption. In the absence of incorporating smoking information into the model, we projected the current trend in a larger extent (model M1T) than the default gradual damping of the impact of the current trend in future periods (model M0T), on the assumption that the current trend will continue into the future. The projected lung cancer pattern in males follows the forecast derived from the default Nordpred drift reduction model (M0T) with adjustment for smoking. The different reactions of the projection models to smoking factors for males and females are related to the different historical patterns of tobacco consumption^{Endnote 42, Endnote 43} and the lag of at least 20 years between a drop in smoking rates and subsequent decrease in cancer incidence rates.