Oral Presentation COSA 2015 ASM

Rare cancers in Queensland: quantifying the impact (#52)

Nathan Dunn 1 , Shoni Colquist 1 , Tracey Guan 1
  1. Queensland Cancer Control Analysis Team, Queensland Health, Brisbane, QLD, Australia

Background

We compared the prevalence of rare, less than common and common cancers in Queensland during 2012 and the potential years of life lost (PYLL) as a result of these cancer diagnoses over the period 2003-2012.

Methods

We aggregated cancer data from Queensland residents into rare cancers, less than common cancers and common cancers.  Cancers were designated as rare or less common (RLC) in accordance with the Rare Cancers Australia Rare Cancers Baseline Report.  Further delineation was obtained by classifying rare cancers as those with an incidence of <6 per 100,000.  Prevalence estimates for these cancer groups among the Queensland population were calculated as at 31 December 2012.  Potential years of life lost (PYLL) was calculated using two different approaches, the life expectancy at birth stratified by remoteness and indigenous status, and the more common PYLL prior to a specified age (75, 80, 85). 

Results

RLC cancers comprised just over one-third (34.5%) of new cancer diagnoses in 2012, whilst being responsible for almost half (47%) of all cancer related deaths during the same year.  The different methodologies of calculating PYLL and varying timeframes for measuring prevalence did not have a major impact on the contribution of RLC to the total cancer burden.  RLC cancers contributed just under 30% of the prevalence burden in each of the 5, 15 and 25 year prevalence estimates.  Between 46% and 48% of PYLL were due to RLC cancers depending on the methodology and age cut-off used.

Conclusions

While individual cancer types show varying patterns of prevalence and PYLL depending on the case fatality rate of each cancer and age-distribution at diagnosis, the collective impact of rare, less than common and common cancers at a state-wide level remained similar whichever outcome measure was chosen.  Contextual information regarding the intended use is key when selecting an outcome measure.