The idea of this study was to explore if we can use serial ctDNA dynamics to assist tumour response to therapy in solid tumours. We primarily hypothesised that changes at the molecular level that we can assess using ctDNA monitoring, or liquid biopsy, can be concordant with changes at the gross level of tumour tissue that can be assessed by radiological imaging. The radiological imaging and molecular assessment of ctDNA can be complementary to each other in situations where the imaging alone cannot be conclusive.
We used a genomically informed approach to quantify different mutations in each of our samples that are collected at different time points using digital droplet PCR. Digital droplet PCR is a new technology that is dependent on the creation of oil-water emulsion droplets in each of which PCR amplification occurs leading to high sensitivity and absolute quantification of the level of mutation in samples. Then we did statistical analysis to see if the changes at the molecular level assessed by quantification of ctDNA are concordant or are reflecting some sort of clinical outcomes at first restaging.
It was interesting in our study to find that in over 200 patients that were included we found that ctDNA progression preceded or co-occurred with the study progressive disease in nearly two-thirds of patients. Moreover, ctDNA dynamic changes were concordant with clinical outcomes as assessed by RECIST imaging at first restaging which means that we can use ctDNA dynamic changes to predict tumour response to therapy in patients with solid tumours.
We had a small cohort of patients treated with immunotherapy in which we validated our results to confirm that it’s not treatment specific. In this cohort of patients it was very interesting to find that in patients with radiological queries to the progression ctDNA dynamics were actually of great value in determining patients who are more free from progression progression versus patients who are in just pre-progression who are likely to continue to benefit from therapy.
Findings from this study can have larger implications in clinical practice because what we are proposing right here is that a simple blood test can be used serially to monitor treatment response to therapy at the molecular level. It’s free from the radiation hazards that occur with radiological imaging and also can detect very many new changes at the molecular level that occur before the real gross changes occur in tumour tissue.
Also, this study showed that ctDNA dynamics are able to identify responders from non-responders and progressors from non-progressors very early in the course of treatment, up to day 21 of therapy, which means that we can very early determine if a patient is going to benefit from therapy or is going to waste his time taking unnecessary medication.
These findings will still need to be validated in a larger cohort and in different tumour types and they will need to be studied in the context of clinical trials. But so far we are very enthusiastic about the results we’ve got because we included a very large number of patients and we included different tumour types and we included different treatment options which means that initially or primarily our results can be encouraging for further studies.