Educational theme | The role of ctDNA in lymphoma management

This article is the first in an educational series exploring the role of circulating tumor DNA in the management of patients with lymphoma.

Cell-free DNA (cfDNA) in the blood primarily originates from apoptosis of hematopoietic cells in healthy individuals. In patients with lymphoma, a proportion the cfDNA originates from tumor cells undergoing apoptosis, designated circulating tumor DNA (ctDNA). The amount of ctDNA varies depending on several factors. More aggressive tumors, tumors of a greater volume and tumors undergoing disease progression as opposed to those in remission, shed greater amounts of ctDNA.¹

Assessment of ctDNA by liquid biopsy is an appealing approach to monitor disease due to its minimally invasive nature. Whilst not currently applied in routine clinical practice, data supporting the use of ctDNA as a tool in lymphoma management is emerging and will be reviewed over this educational series.

Davide Rossi recently spoke with the Lymphoma Hub on “why should we use ctDNA assessments in lymphoma diagnostics in the future?”

During the 62nd American Society of Hematology (ASH) Annual Meeting and Exposition, several talks were presented regarding the prognostic value of ctDNA in patients with diffuse large B-cell lymphoma (DLBCL). They were previewed in our pre-ASH summary article here. In the current article we would like to delve deeper into the results presented during two of these sessions.

Prognostic value of ctDNA in autologous stem cell graft and posttransplant plasma samples among patients with DLBCL

Although autologous stem cell transplantation may be curative for some patients with DLBCL, a significant number of patients relapse, and biomarkers are required to identify patients who are most at risk. Reid Merryman and colleagues² investigated whether ctDNA-based minimal residual disease assessment in apheresis stem cell (ASC) and posttransplant plasma samples could predict relapse in patients with DLBCL. Immunoglobulin-based next-generation sequencing (NGS) was used to identify ctDNA within the samples.

Of the 96 patients who had ASC samples, 24% were ctDNA-positive. For these patients, the 5-year progression-free survival (PFS) was only 13% (95% CI, 3−30%), compared with 52% (95% CI, 39−63%) for patients who were ctDNA-negative. The 5-year cumulative relapse rate was 83% (95% CI, 66−99%) for the ctDNA-positive group and 39% (95% CI, 28−51%) for the ctDNA-negative group. ASC ctDNA was significantly associated with PFS (HR 2.8; p < 0.01) and was the only factor shown to be predictive following multivariate analysis (HR 2.5; p = 0.002). A similar association was seen between ctDNA positivity and overall survival (OS). Patients who were ctDNA-positive had a lower 5-year OS (51% [95% CI, 29%−69%]), compared with patients who were ctDNA-negative (66% [95% CI, 54%−75%]; p = 0.048).

Posttransplant plasma samples from 56 patients were also examined for ctDNA content and positive samples were found to be associated with increased risk of relapse. Overall, 21 patients (38%) had a median of two ctDNA-positive posttransplant samples; 86% of these patients relapsed.

We asked Reid Merryman, “can ctDNA help inform on the suitability of autologous stem cell transplantation (auto-SCT) in patients with DLBCL?” in this Lymphoma Hub podcast.

Risk profiling of relapsed/refractory DLBCL patients by measuring ctDNA

In another report from ASH 2020, Alex Herrera et al³ analyzed whether ctDNA could identify patients at a higher risk of relapse in GO29365, a phase Ib/II study (NCT02257567), of relapsed/refractory DLBCL treated with bendamustine and rituximab +/− polatuzumab. Plasma samples were available at baseline for 43 patients; however, eight patients were excluded from the ctDNA analysis due to lack of paired germline data. The remaining 35 patients included in the biomarker-evaluable population (BEP) were positive for ctDNA at baseline. The characteristics and the PFS between the BEP group and the 80 patients included in the intention-to-treat population were found to be similar; the BEP group was, therefore, a representative sample of the whole cohort.

An association between high levels of ctDNA at baseline and a poor prognosis was noted with respect to both PFS and OS. When stratifying patients by median ctDNA levels, the unadjusted HR for PFS was 0.25 (95% CI, 0.07–0.93), and for OS was 0.23 (95% CI, 0.10−0.52). The prognostic value of ctDNA at baseline was retained in multivariate analysis when adjusting for treatment, IPI, and LDH > upper limit of normal (PFS, HR 0.24 [95% CI, 0.07−0.81]; OS, HR 0.30 [95% CI, 0.09–0.97]).

A change in ctDNA levels over time was correlated with treatment response. Patients with a complete response (CR) at the end of treatment (EOT) (n = 13) demonstrated a significantly greater reduction in ctDNA levels from baseline to EOT compared with patients who did not reach CR (p = 0.00097). At EOT, four of 25 patients were negative for ctDNA and all achieved a CR. The remaining 21 patients were positive for ctDNA. No association was found between a reduction in ctDNA and PFS in this cohort, although the small sample size was noted.

When Alex Herrera spoke to the Lymphoma Hub we asked, “does ctDNA add value to current prognostic markers for identifying high-risk patients?”

Conclusion

The use of liquid biopsy to measure ctDNA offers a noninvasive method to predict relapse in patients with lymphoma. This technique has shown great sensitivity in a number of small studies. While larger studies using standardized methodology are still required, there is great potential for this technique to benefit both clinicians and patients. For patients, this could mean fewer invasive procedures and for clinicians, the chance to intervene earlier if signs of relapse are indicated.

For other articles in this theme click on the links below:

Monitoring treatment response in lymphoma using cfDNA

CSF analysis of ctDNA for lymphoma with CNS involvement

5hmC profiles of cfDNA predict R-CHOP treatment response in patients with DLBCL