e-Poster Presentation Clinical Oncology Society of Australia Annual Scientific Meeting 2021

Using patient reported outcome measures to determine the role of the microbiome in cancer treatment-induced diarrhoea (#340)

Kate Secombe 1 2 , Imogen Ball 2 , Courtney Subramaniam 2 , Ghanyah Al-Qadami 2 , Joanne Bowen 2 , Hannah Rose Wardill 2 3
  1. The University of Queensland Diamantina Institute, University of Queensland, Woolloongabba, QLD, Australia
  2. School of Biomedicine, University of Adelaide, Adelaide, SA, Australia
  3. Precision Medicine (Cancer) Theme, South Australian Health and Medical Research Institute, Adelaide, SA, Australia

Aims: Diarrhoea is a commonly occurring side effect of many cancer treatments, however it is unclear why some people develop severe diarrhoea and others do not, independent of cancer or treatment type. A growing body of research suggests the treatment-naïve gut microbiome composition may be used as a predictor of diarrhoea development. However, this has not previously been assessed in a prospective, longitudinal cohort, using patient reported outcome measures (PROMs).

Methods: We have developed the PREDiCT (Personalised Risk Evaluation During Cancer Treatment) study in three South Australian oncology centres. To date, 45 participants have been recruited. Stool samples are collected longitudinally throughout the first two treatment cycles (including pre-therapy baseline) for analysis using 16S rRNA gene sequencing. PROMs assessing diarrhoea (FACIT-D) and other treatment side-effects (C-SAS) are assessed via validated surveys at time points matched to stool sampling.

Results: The microbiome composition and FACIT-D data of fifteen participants with colorectal, breast or cervical cancer undergoing chemotherapy have been analysed. A clinically important threshold in diarrhoea scores was determined, and participants were then stratified into ‘diarrhoea’ and ‘no diarrhoea’ groups. There were no significant differences in pre-treatment microbiome composition between these two groups. Analysis of mid-treatment microbiome composition is ongoing.

Conclusions: In this cohort, participants were unable to be stratified based on diarrhoea occurrence using microbiome composition prior to treatment, possibly due to small sample size and cancer type heterogeneity. However, the study so far demonstrates that using PROMs to assess treatment-related diarrhoea occurrence is feasible. Additionally, there is a sound rationale for future development of the PREDiCT study, with an increased sample size enabling more sensitive modelling to be achieved through the inclusion other co-variants (including demographic/treatment-related and host genetics) which have been prospectively collected from all participants.