Limitations of Monitoring Disease Progression Using Circulating Tumor DNA in Lymphoma: An Example from Primary Cutaneous DLBCL Leg-type

Chemotherapy-refractory diffuse large B-cell lymphoma (DLBCL) remains a significant clinical problem. The ability to predict patients likely to have poor outcomes with conventional therapies may facilitate rational use of alternative, targeted treatment approaches before fulminant relapse. The utility of circulating tumor DNA (ctDNA) in DLBCL is currently being investigated. Initial studies have reported high levels of sensitivity for detection of residual disease, outperforming imaging techniques such as 18FDG-PET/CT scans.1 For example, Alizadeh et al recently demonstrated the feasibility of CAncer Personalized Profiling by deep Sequencing (CAPP-Seq) approach in DLBCL; the mean elapsed time between the first ctDNA-positive time point and radiological relapse was 188 days.2 Kurtz et al detected pretreatment ctDNA in 98% of patients with DLBCL receiving treatment with frontline or salvage immunochemotherapy.3 Early and major molecular responses after 1 and 2 cycles of treatment, resulted in superior outcome at 24 months. However, only 9% and 23% of patients within these validation sets had stage I and II disease, respectively. Limitations to the use of ctDNA to detect early stage malignancy have been identified; for example, in lung, where low-volume disease cannot be reliably detected using mutation profiles in ctDNA.4 Furthermore, some specific subtypes of disease (lung adenocarcinoma) do not shed detectable ctDNA into the peripheral blood.5