Continuous Increasing Prevalence of Chronic Lymphocytic Leukemia (CLL) with an Estimated Future Rise—a Nationwide Population-Based Study from Sweden

Rationale

Data on the prevalence of chronic lymphocytic leukemia (CLL) are scarce, and reliable population-based data as well as estimates of future changes in prevalence are needed

Aims

To investigate previous and current prevalence of CLL in Sweden, the time it takes for new treatments to be implemented in clinical practice and to estimate the future prevalence of CLL.

Patients and Methods

Individuals in Sweden diagnosed with CLL (ICD7 and ICD9 204.1 and ICD10 C91.1) during 1958-2015 were identified through the National Cancer Registry. Background population sizes, prediction of future population sizes, death rates and data on emigration and immigration were obtained from Statistics Sweden. Clinical data on all cases diagnosed 2007-2015 (n=4742), including type of treatment for those requiring therapy (n=1459), were obtained from the Swedish National CLL-registry to which all healthcare units (n=73) diagnosing and treating CLL send reports.

The number of prevalent cases at the end of each year was calculated from the total number of individuals diagnosed with CLL still alive and living in Sweden. Future age-specific incidence rates were assumed to be equal to the average age-specific incidence rates in the previous ten years. To estimate the proportion of individuals still alive many years into the future, estimates of relative survival along with expected mortality rates were calculated. These estimates were used in conjunction with predictions of future expected background mortality to predict overall survival among prevalent cases. Using a flexible parametric model (Lambert, Wilkes et al. 2017) with age at diagnosis included as a restricted cubic spline, relative survival was estimated for all ages.

To create a "best case" scenario of future prevalence, a sensitivity analysis was performed assuming an increase in survival of newly diagnosed patients based on hazard ratios from randomized trials comparing new treatments with old. From the Swedish National CLL-registry we gathered data on treatments used and made assessments within three age groups (<65 years, 65-79 years and >80 years) on the proportion of patients that would benefit from a change to a more efficient treatment.

Results

In 2015, the age-standardized incidence of CLL in Sweden was approximately 3/100 000 among males and 1.5-2/100 000 among females. This level had been stable since 1980.

During the period 2000 to 2015 there was a continuous improvement in survival. In parallel there was a 42% increase in point prevalence from 33.3/100 000 to 47.4/100 000, corresponding to a 58% increase in the absolute number of prevalent cases from 2954 to 4668 (fig. 1A).

Novel treatments were implemented in <5 years in general practice (fig. 1B).

Using this information, we calculated a "best-case" scenario, applying survival figures based on recent published studies on novel treatments, and assuming that the maximum proportion of patients receiving the novel drugs would be reached within 5 years.

Assuming unchanged relative survival, we estimated a 15% rise in the point prevalence to 54.6/100 000 in 2025 and a 26% rise to 59.5/100 000 in 2035, corresponding to an increase in the absolute number of cases with 28% (n=5980) and 49% (n=6938) during the same period of time.

Based on the "best case" scenario, we estimated a 21% rise in the point prevalence to 57.3/100 000 in 2025 and a 45% rise to 68.9/100 000 in 2035. This corresponding to an increase in the absolute number of cases during the same period with 35% (n=6280) and 72% (n=8016), respectively. (fig. 1C)

Conclusion

Using data from well-established national, population-based registries in Sweden, focusing on the time period 2000-2015, we found that there has been a continuous increase in the prevalence of CLL from 33.3/100 000 (2000) to 47.4/100 000 (2015). We also describe a fast implementation of new treatments. In parallel we observe a striking increase in survival. Using a model in which we assume further improvements in survival due to the introduction of novel and more efficient treatments, we estimate that the absolute number of cases may increase with as much as 72% in 20 years. These findings have important implications for health economy and planning.

Acknowledgments

The Swedish Regional Cancer Centers and all hematologists reporting patients to the National CLL Registry. The Swedish CLL-group for their valuable support.

Funding

This work was supported by funding from the Swedish CLL-group.

View largeDownload slide

View largeDownload slide

Close modal