Introduction:
Prostate cancer is a disease of the elderly. Higher co-morbidities, functional and mental decline are associated more frequently in this population. These have greater impact on the ability of elderly patients to tolerate and receive optimal treatment for their disease. Numerous studies have shown the importance of screening elderly patients via Comprehensive Geriatric Assessment (CGA) to allow risk stratification of prostate patients and to enable identification of areas that could be optimised prior to commencement of treatment. Little has been done previously in Australia to screen prostate patients via CGA to guide treatment.

Methods:
Retrospective observational study on patients aged ≥70, with prostate cancer treated at Royal Adelaide Hospital and Modbury Hospital from June 2008 - June 2013.Abbreviated CGA was used as geriatric assessment at baseline. The study population was stratified into 3 groups (‘fit’, ’vulnerable’, ‘frail’) after completion of CGA. The survival time of patients from each group were investigated.

Results:
85 patients were identified. Mean age was 78.86 years. 33.3% were classified as ‘fit’, 46.2% ‘vulnerable’, 20.5% ‘frail’. 52 patients deemed castrate resistant received Docetaxel; 14 (53.85%) from ‘fit’, 19 (52.8%) from ‘vulnerable’, and 5(31.3%) from ‘frail’ group. Survival was improved for patients in ‘fit’ and ‘vulnerable’ group with treatment from Docetaxel, 18.2 vs 14.4 months, 16 vs 6.2 months, respectively. This was not seen in the ‘frail’ group, 5.8 vs 8.9 months with and without Docextaxel. 5 year survival rate for ‘fit’ was 39.5%, ‘vulnerable’ 22.8% and ‘frail’ 12.5%. Impaired level of functioning (IADLS, 59%, ADLs 47.4%) and lack of psychological support (39.7%) were identified as the top three ‘geriatric issues’ of concern in all 3 groups.

Conclusion:
The study confirms the importance of CGA in assessing elderly with cancer. The CGA tool has strong prognostic and predictive value. Patients in the frail group did not benefit from the addition of chemotherapy.