Abstract 1891
Background
Elderly patients (pts) are at increased risk of developing chemotherapy (CTx) related toxicities. There are several prediction tools including the Cancer and Aging Research Group (CARG) chemotoxicity calculator (CTC) but this has not been validated in the Asian population. The objective of our study is to identify predictors of G3-5 CTx toxicities and evaluate the utility of the CTC in older Asian adults.
Methods
We enrolled cancer pts aged ≥ 70 years, requiring outpatient treatment with chemotherapy at the National University Cancer Institute, Singapore. A comprehensive geriatric assessment was performed and baseline cancer characteristics, including the 11 variables in the CTC were collected prior to the initiation of CTx. Primary treating oncologists were asked to give an estimated likelihood of CTx related toxicities. Pts were then followed up to 3 months after completion of treatment and CTx toxicities were recorded.
Results
Amongst the 131 pts (mean age: 75 years; range: 70 – 89), their CA diagnoses include colorectal (CRC) (36; 27.5%), lung (27 ; 20.6% ), non-CRC gastrointestinal (17; 13%) , breast (14 ; 10.7%), genitourinary (10 ;7.6%), head and neck (10 ; 7.6%), gynaecological (8; 6.1% ) and other (9; 6.9%) cancers. 54 pts (41%) received CTx with curative intent, and 77 (59%) with palliative intent. The incidence of G3-5 toxicities was 58% (76). The most common toxicities were neutropenia (38 ; 29%) and anaemia (26; 20%) and there was 1 mortality from CTx related pneumonitis. In the multivariable analysis, the factors associated with the incidence of CTx related toxicities were i) Age >72 years, ii) GI/GU cancer, iii) Haemoglobin < 10g/dL (Female) ; <11g/dL (Male), iv) Limited ability to walk 1 block, iv) Limited social support and v) Disease interference with social activity. Based on this new model comprising the above 5 variables, the area under the receiver operating characteristic curve (ROC) is 0.776, whereas that for the CTC and oncologists’ prediction of CTx related toxicities were 0.765 and 0.594 respectively.
Conclusions
Our new model for predicting CTx related toxicities appears to be comparable to the CTC. In our follow up study, we hope to further validate this model’s utility in predicting CTx related toxicities in our elderly cancer population.
Clinical trial identification
Legal entity responsible for the study
Angela Pang.
Funding
National Medical Research Council (Singapore)
Editorial Acknowledgement
Disclosure
All authors have declared no conflicts of interest.