The study presented at SIOG we were looking at the association between epigenetic age and the frailty index in older women with early breast cancer prior to getting chemotherapy.

What was the study design?

This design was a secondary analysis, a cross-sectional study of the HOPE cohort, the Hurria Older Patients with breast cancer cohort. This was a cohort of about 500 women who were 65 and older with a diagnosis of stage 1 through 3 early breast cancer who were slated to get chemotherapy prior to treatment. We looked at different measures of biological and functional aging in this patient population at this one timepoint.

What were the results of this study?

We found that there was an association between the measures of biological aging and measures of functional aging in this patient population. The measures of biological aging that we used were epigenetic clocks. These are algorithms that measure the DNA methylation patterns that we see as we age and these clocks are developed to compare biological age with chronological age. We used three generations of epigenetic clocks for this study: the first generation were the Horvath and Hannum clocks, the second generation was the GrimAge and PhenoAge clocks and the third generation was the DunedinPACE, which is a measure of rate of biological aging measured in biological years over chronological years.

To look at the first and second generation clocks we also calculated something called the epigenetic age acceleration. So this is the residual that results from regression of the biological age as determined by the clocks over the chronological age of this patient population.

We then looked at the association between these epigenetic age acceleration markers in these women at this timepoint and compared it to our functional measures. Functional measures, the main outcome was the deficit-accumulation frailty index which is a measure of frailty. This takes a compilation of 50 different items, multidimensional physical, social and cognitive function measures that we have collected on these patients and generated a score. The score reflects the frailty status depending on prior studies showing different cut-off points to define robust pre-frail and frail patients.

So we found that among our patient population about 25% were pre-frail or frail and these pre-frail or frail patients had significantly higher rates of epigenetic aging, age acceleration, compared with those who were robust, looking at the different epigenetic clocks. In particular, the second and third generation clocks were the ones that showed statistical significance so those were the clocks that demonstrated those who were pre-frail or frail had increased rates of epigenetic age acceleration compared to those who were robust. One of the clocks, for example the PhenoAge clock, showed that these women were on average… the women who were pre-frail and frail were on average 4½ years older than those who were robust biologically and aged at about a 10% faster pace of aging.

So we showed this association but we also saw that there was significant heterogeneity in the groups which means that even among those who are robust there were still some who did not have any evidence of epigenetic age acceleration and among those who were frail and pre-frail there were those who had younger biological age or no age acceleration. So this is not a perfect correlation with frailty, there are definitely discrepancies in these different measures of aging with biological and functional aging.

What is the significance of these results?

The clinical significance for these results really is very preliminary evidence that measures of biological aging may give us additional information on top of our measures of functional status and frailty in the clinic. This is still a way off from potential clinical translation. I think it sets the stage for future work where we look at how does epigenetic age predict outcomes in our patients in terms of, for example, treatment toxicity or hospitalisation or tolerability of treatment and even cancer outcomes, potentially, in the future. Until these studies are done we can’t really compare to how they’re better or how they supplement our traditional measures of function and functional aging. Can we add to our traditional measures to better refine the treatment recommendations for our older adults who may be more susceptible to treatment toxicities and outcomes and adverse outcomes?