Although the field of circulating tumor DNA for DLBCL is young and clinical validation is needed, this technology is sure to become part of routine clinical decision-making.— Mark Roschewski, MD, and Wyndham H. Wilson, MD, PhD
Technologic advances for detecting and analyzing cell-free circulating tumor DNA (ctDNA) from peripheral blood offer a precision method for monitoring diffuse large B-cell lymphoma (DLBCL). Although most patients with DLBCL are cured with initial therapy, those who are not cured have a poor prognosis and represent an unmet medical need. Precision monitoring of ctDNA has generated interest because early detection of treatment failure and targetable mutations may significantly improve patient outcomes and represents an important research tool. Previous efforts to identify treatment failure have utilized radiographic imaging with computed tomography (CT) and positron-emission tomography (PET) scans, but they appear to be of marginal clinical utility and are not cost-effective.1-3
Every DLBCL tumor harbors a unique gene rearrangement within the VDJ region of the immunoglobulin receptors, making ctDNA encoding clonal VDJ sequences a tumor-specific marker.4 Two independent studies demonstrated that next-generation sequencing–based methods can detect ctDNA encoding VDJ in patients with DLBCL months prior to clinical relapse.5,6
These results have emboldened a search for additional applications, including assessing genotypic DNA shed from tumor cells. As reviewed in this issue of The ASCO Post, Scherer and colleagues reported impressive results using a novel sequencing-based method of ctDNA profiling that detects highly relevant lymphoma-specific genetic aberrations in addition to clonal VDJ sequences.7-9 They showed potential applications of a “liquid biopsy” that can interrogate important tumor biology including tumor heterogeneity, clonal evolution, and mechanisms of resistance. Indeed, these findings justify further efforts to incorporate precision ctDNA monitoring into the treatment of DLBCL.
What Challenges Can ctDNA Overcome?
The primary advantage of monitoring ctDNA over radiographic imaging is the ability to assess disease at the molecular level. Before radiologic imaging can reliably detect disease, the tumor burden must reach a minimum of nearly 1 billion cancer cells. In contrast, current ctDNA methods are capable of detecting lymphoma molecules at a threshold of 1 × 106 cellular equivalents.10 Early detection of disease provides an opportunity for earlier salvage therapy. Although it is unclear whether early institution of treatment will improve outcomes, it has rational appeal because these patients are potentially curable.
Molecular measurement of ctDNA may also re-define our definition of disease remission. Currently, up to 15% of patients with DLBCL who are PET-negative after standard therapy will experience disease relapse, indicating the presence of microscopic disease.11 In addition to serving as a molecular marker of disease, ctDNA may also provide actionable molecular information as a liquid biopsy to complement or replace invasive tissue biopsies.12
The considerable variation in the molecular biology and clonal heterogeneity of DLBCL provides a compelling rationale for liquid biopsies. Biopsies of single tumor sites are confounded by sampling error and hence do not adequately capture the tumor heterogeneity. Furthermore, clonal evolution over time and in the presence of therapeutic selective pressure can significantly alter the tumor biology and treatment sensitivity.13,14 Currently, clinical decision algorithms are based almost entirely on the baseline tumor and are mostly agnostic at relapse. Thus, ctDNA profiling that can serially and noninvasively provide genetic information will establish an important new paradigm for personalized medicine in DLBCL.15
Novel Study Findings
Prior to the study by Scherer and colleagues, there was little published information on ctDNA mutational profiling in DLBCL other than a small series of 12 patients.16 These Stanford investigators greatly expanded our knowledge by applying a novel targeted sequencing panel specifically designed for relevance in DLBCL known as cancer personal profiling by deep sequencing (CAPP-Seq).8,9
CAPP-Seq can detect immunoglobulin receptor sequences; breakpoints in genes such as BCL2, BCL6, and MYC; and other recurrent single nucleotide variants and insertions/deletions considered relevant, based on published results from whole-exome and whole-genome sequencing studies in DLBCL.17-20 The authors applied CAPP-Seq profiling to both lymphoma tissue and plasma samples in 50 patients with de novo DLBCL, 19 patients with DLBCL that had transformed from an indolent lymphoma, and 7 patients with primary central nervous system lymphoma.
The first notable finding was that CAPP-Seq had universal sensitivity with detectable ctDNA in all patients with pretreatment samples (N = 45) and few false-positive results. Moreover, the quantitative level of pretreatment ctDNA in plasma was prognostic. Similar to published results, the plasma ctDNA concentration correlated with clinical measures of tumor burden, such as serum lactate dehydrogenase level, stage, and the International Prognostic Index. It is important to note that on multivariate analysis, the pretreatment concentrations of ctDNA correlated with progression-free survival. Further studies are needed to explore the biologic relevance of quantitative ctDNA to measures of tumor proliferation, oncogenic signaling, and as a surrogate for immune evasion.
A second key finding was the demonstration that genotyping of plasma ctDNA performed well compared with analysis of tumor biopsies, particularly when the ctDNA concentration was high. The majority of single nucleotide variants in driver genes found in tumor tissue, and 89% of relevant translocations identified by fluorescent in situ hybridization, were detectable in the plasma. At the patient level, 39 of 45 patients (87%) had a tumor-derived biomarker identified in their pretreatment plasma. These results compare favorably with results reported with ctDNA for VDJ sequences, but the authors also described numerous cases identified by CAPP-Seq that were not identified by ctDNA for VDJ sequences only. It is certainly possible that a combination of VDJ and mutational analysis will be more sensitive than VDJ alone, but further study in larger cohorts is needed to compare these methods in various clinical settings.
The most notable finding for the future of precision medicine is the ability to reliably perform “biopsy-free genotyping.” When the allele fraction was over 20%, almost all of the single nucleotide variants in driver genes were identified in the plasma, allowing for serial sampling and assessment of clonal evolution. Specific applications of noninvasive genotyping were highlighted by the authors, including a very high detection rate of BCL2, BCL6, and MYC translocations, as well as an ability to determine the DLBCL molecular subtype according to the cell-of-origin classification.
The authors showed compelling examples of CAPP-Seq’s ability to serially study clonal evolution, including the detection of treatment-emergent mutations that confer resistance to targeted therapy. In one striking example, they described a case in which two distinct point mutations in the binding site of ibrutinib (Imbruvica) emerged after the patient was started on ibrutinib. Another application was highlighted by analyses of serial plasma samples from patients with indolent lymphomas who developed histologic transformation to DLBCL. In these cases, the authors found that patients who developed transformation, compared with those who did not, had higher ctDNA concentrations and a greater number of unique mutations in plasma obtained before transformation. If validated, such information could help predict patients at highest risk for transformation and identify potentially targetable mutations.
Immediate Impact and Future Directions
The study by Scherer et al offers multiple potential applications of ctDNA monitoring for the precision management of DLBCL. Indeed, plasma-based genotyping by CAPP-Seq provides the first experimental evidence that dynamic biologic processes such as clonal evolution can be directly assessed in real time without the need for tissue biopsies. Although it is important to note that the field of ctDNA for DLBCL is young and clinical validation is needed, this technology is sure to become part of routine clinical decision-making. Further advances are expected, and the technology will continue to grow in sensitivity and in the breadth of biologic endpoints that can be measured. ■
Disclosure: Drs. Roschewski and Wilson reported no potential conflicts of interest.
3. El-Galaly T, Prakash V, Christiansen I, et al: Efficacy of routine surveillance with positron emission tomography/computed tomography in aggressive non-Hodgkin lymphoma in complete remission: Status in a single center. Leuk Lymphoma 52:597-603, 2011.
5. Roschewski M, Dunleavy K, Pittaluga S, et al: Circulating tumour DNA and CT monitoring in patients with untreated diffuse large B-cell lymphoma: A correlative biomarker study. Lancet Oncol 16:541-549, 2015.
11. Adams HJ, Nievelstein RA, Kwee TC: Prognostic value of complete remission status at end-of-treatment FDG-PET in R-CHOP-treated diffuse large B-cell lymphoma: Systematic review and meta-analysis. Br J Haematol 170:185-191, 2015.
14. Melchardt T, Hufnagl C, Weinstock DM, et al: Clonal evolution in relapsed and refractory diffuse large B-cell lymphoma is characterized by high dynamics of subclones. Oncotarget 7:51494-51502, 2016.
16. Bohers E, Viailly PJ, Dubois S, et al: Somatic mutations of cell-free circulating DNA detected by next-generation sequencing reflect the genetic changes in both germinal center B-cell-like and activated B-cell-like diffuse large B-cell lymphomas at the time of diagnosis. Haematologica 100:e280-e284, 2015.
19. Lohr JG, Stojanov P, Lawrence MS, et al: Discovery and prioritization of somatic mutations in diffuse large B-cell lymphoma by whole-exome sequencing. Proc Natl Acad Sci U S A 109:3879-3884, 2012.