Educational content on VJOncology is intended for healthcare professionals only. By visiting this website and accessing this information you confirm that you are a healthcare professional.

Share this video  

ASCO 2022 | ctDNA in predicting relapse in solid tumors treated with curative intent therapy

Abhenil Mittal, MD, DM, Princess Margaret Cancer Center, Toronto, Canada, discusses results from a study quantifying the predictive and discriminatory accuracy of circulating tumor DNA (ctDNA). ctDNA shows high prognostic accuracy, however, its low sensitivity and suboptimal specificity highlight the need for optimization before it can be implemented in routine clinical practice. This interview took place at the American Society of Clinical Oncology (ASCO) 2022 Annual Meeting in Chicago, IL.

Transcript (edited for clarity)

I’ve been working there for the past eight months and we do a lot of translational work as well as a lot of health services work and clinical trials at Princess Margaret. It’s one of the largest institutions in Canada. I had two presentations at ASCO this year, both were poster presentations. One was yesterday in the developmental therapeutics and immunotherapy session. This was role of ctDNA in predicting relapse in solid tumors, which have been treated with curative-intent therapy, meta-analysis of published literature...

I’ve been working there for the past eight months and we do a lot of translational work as well as a lot of health services work and clinical trials at Princess Margaret. It’s one of the largest institutions in Canada. I had two presentations at ASCO this year, both were poster presentations. One was yesterday in the developmental therapeutics and immunotherapy session. This was role of ctDNA in predicting relapse in solid tumors, which have been treated with curative-intent therapy, meta-analysis of published literature. There has been recent hype and talk about the role of circulating tumor DNA as a measure of minimal residual disease in patients with solid tumors who have been treated with curative-intent treatment. However, we are yet to optimally understand how and when to utilize this extremely exciting tool.

So we try to look at all the published literature, which has been there so far in different tumor sites and look at various time points at which ctDNA has been measured in them and see what the sensitivity and specificity is of this tool. We pulled that data in a meta-analysis and then we did regression to understand what factors, both disease and treatment-related, play a role in defining the sensitivity and specificity of ctDNA. To briefly summarize the results, we found about 38 studies, which have looked at it over the past five years, which is a huge number. We looked at sensitivity and specificity at both a landmark time point, which is a single time point post curative treatment where ctDNA was done, and we looked at surveillance time points that is at multiple time points post treatment.

We found that the sensitivity was low; at a 60% at a landmark time point. Specificity was around 90%, which means that we were likely to find 40% false positives or 10% false negatives. In other words, maybe over treat or under treat some patients if we were just to use ctDNA as the stand-alone tool. If we were to use the… this was affected by adjuvant treatment; adjuvant chemotherapy most of it. And if it were to be done post adjuvant chemotherapy, the sensitivity was slightly better. However, then that does not inform adjuvant chemotherapy decisions, which is where it is being studied mostly right now in clinical trials. If we are using it as a surveillance tool, the sensitivity was better with 82% and specificity of around 94%. However, that informs decisions mostly for detecting early relapse and would inform future clinical trials as to if we could intervene earlier and whether that would affect survival would have to be proven in randomized trials. We have gone that route before and treated asymptomatic patients based on a marker. The best evidence being in CAOV, and have burned our fingers before.

We would need to be careful in terms of designing future studies and taking this… and we should just keep this data in mind and apply this in context while designing and looking at future studies.

Read more...