Large cell lymphomas are curable, but aggressive. Despite that, some 30% to 40% of patients with diffuse large B-cell lymphoma (DLBCL) will develop relapsing/refractory disease.¹ To ensure a cure, physicians typically use CT and positron emission tomography-computed
tomography (PET-CT) scans to look for signs of minimum residual disease (MRD) when treatment is completed.

“If you can cure something by treatment, and you’re detecting circulating tumor DNA at the end of treatment, then that can be useful to know, because that lends itself to a consideration that this lymphoma can come back, and what treatments do you plan from there?” Rahul Lakhotia, MD, who is an assistant research physician in the lymphoid malignancies branch of the National Cancer Institute (NCI), said during an interview with Targeted Therapies in Oncology.

But CT and PET-CT aren’t always up to the task. For end-of-treatment assessment in aggressive non-Hodgkin lymphoma (NHL), PET-CT has a negative predictive value of 80% to 100%, but a positive predictive value of 50% to 100%. The method is generally used in the decision whether to intensify therapy, but PET-CT is not sensitive enough to detect MRD.
Instead, invasive biopsies or follow-up scans are recommended.²

Circulating tumor DNA (ctDNA) has the potential to address these problems in a minimally invasive way. Dying and proliferating cells release cell-free DNA (cfDNA) that makes its way to peripheral blood, where it can be captured with a simple blood test. ctDNA can be measured as a proportion of overall cfDNA, and this is being investigated in a variety of lymphomas including aggressive lymphomas such as DLBCL and Hodgkin lymphoma, according to Lakhotia.

It has the potential to dramatically change treatment of large cell lymphoma, but a range of challenges remain to be solved before it can enter widespread clinical application.

A Moving Target

Large cell lymphomas encompass a broad range of clinical features and genetic alterations. That heterogeneity has led to a demand for methods
to provide a prognosis and monitor the course of the disease, but existing techniques are limited. Current prognostic methods rely on disease characteristics at diagnosis and don’t incorporate treatment response or disease evolution. In fact, lymphomas and other cancers are believed to undergo clonal evolution, in which genetic diversity arises and a single clone or group of clones grows to dominate the tumor.³

Information about these clones and their mutational profiles could guide treatment choice or inform prognosis, but current clinically available tools such as CT and PET scans can’t capture these characteristics. Genomic profiling can capture clonal evolution, but it requires invasive biopsies and cannot capture spatial differences in tumor composition, and it does not reveal clonal evolution following treatment.

ctDNA can pick up any tumor DNA and analyze it for the presence of a wide range of mutations, but the field is still struggling with technical challenges. ctDNA has been shown to have prognostic significance, both at diagnosis and during treatment. A study also showed that ctDNA can add to the prognostic information provided by PET scans and other prognostic factors such as international prognostic index (IPI) in DLBCL.⁴ However, important questions remain prior to its clinical application, including the best timing for sample collection, as well as the best assay method. “It is not yet standardized enough to use it as a clinical test,” Lakhotia said.

Technical Challenges

For MRD detection, the most common ctDNA methods are polymerase chain reaction (PCR) and next-generation sequencing (NGS). Both face challenges. ctDNA is present in small amounts of total cell-free DNA, typically less than 0.5%, and a test may be looking for a wide range of potential mutations. A useful test must be able to make use of as much cfDNA as possible and maintain low levels of background error.

PCR-based tests require short stretches of DNA, called primers, that target specific, known mutations. Creation of those primers can be costly and time-consuming. On the other hand, NGS can detect multiple mutations in a single sample, without requiring patient-specific primers, giving it a wider range of applications. "It doesn't need specific probes for each individual patient, and in diffuse large B-cell lymphoma, where there is not usually 1 specific genomic abnormality, these assays can still be utilized. I think going forward, for diffuse large B-cell lymphoma, specifically, I think next-generation sequencing is probably going to be more likely to be used compared to PCR-based methods for those reasons,” Lakhotia said.

There are 2 potential avenues to address challenges presented by low levels of tumor-derived cfDNA. One is to increase the amount of plasma drawn from the patient, but this has practical limitations in clinical care. Another approach is to sequence multiple recurrently mutated genes in diffuse large B-cell lymphoma, which increases the available number of DNA fragments that could bear useful information. One such method, called cancer personalized profiling by deep sequencing (CAPP-seq), has been most extensively studied in DLBCL. CAPP-seq suffers from high background error rate, but this can be improved using error suppression method. “Such approaches to improving the sensitivity for MRD detection represent the next key advance in liquid biopsies for lymphoma, allowing disease detection at the time of lowest disease burden, such as the end of induction therapy,” wrote the authors of a recent review.⁵

These methods also have the potential to provide molecular genotyping. Gene expression profiles have already identified the molecular cell of origin in DLBCL and have also identified mutations in follicular lymphoma, chronic lymphocytic leukemia, and subtypes of DLBCL that can inform prognosis. ctDNA analysis has also revealed targetable EZH2 mutations in follicular lymphoma.

Slow Road to the Clinic

The data from ctDNA studies are tantalizing, but it’s important to demonstrate that they can be used to guide therapy and improve outcomes in prospective studies, according to Lakhotia. There are multiple labs, both in academic institutes as well as pharmaceutical companies, that are studying circulating tumor DNA in their clinical trials and developing those further. There's quite a bit of push to understand its implications in different lymphomas, and it’s probably the most advanced in diffuse large B-cell lymphoma, because that’s where it started, and I think the next 5 or 6 years will probably give us a lot more information about it,” Lakhotia said.

Indeed, the NCI enters almost all of its patients into clinical trials, and ctDNA is now incorporated as a research assay in its lymphoma trials to better understand its potential as a biomarker, Lakhotia added.

To truly be useful, ctDNA must lead to a change in outcomes. That leads to 2 needs. “One, we need a good assay, which we are in the process of developing in circulating tumor DNA,” Lakhotia said. “But what we also need is a treatment that if given early will impact the outcomes afterward. That has not yet been proven. I think future studies will [examine], especially for patients whose ctDNA either remains high or detectable during treatment or after treatment, if changing therapy or having early intervention makes a difference in the real-world outcomes. That would be one of the biggest things that moves this forward.”

Previous studies have suggested the value of ctDNA. A study⁵ by NCI researchers used sequencing to detect ctDNA in retrospectively obtained plasma of patients with DLBCL who had undergone treatment between 1993 and 2013. They found patients with detectable ctDNA during surveillance after therapy were 228 times more likely to experience relapse (P <.0001).

Another group at Stanford evaluated ctDNA analysis among patients with DLBCL at 6 centers in an effort to identify a signature of response to rituximab in combination with cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP).⁶ The study authors noted that the findings could be useful for clinical trial design to intensify therapies in patients whose ctDNA does not change adequately, and also potentially in clinical practice to aid in prognosis. Lakhotia agrees with that sentiment. "Early during treatment, ctDNA can identify patients who are at higher risk of progression after treatment," he said.

Other recent studies have identified additional molecular subtypes of DLBCL and ctDNA seems to hold up well as a prognostic tool. In an abstract at the 17th International Conference on Malignant Lymphoma in June 2023, researchers used ctDNA and genomic DNA isolated from lymph nodes in newly diagnosed and homogeneously treated patients with DLBCL treated with R-CHOP.⁷ The new researchers found that a greater than 2.5-log reduction in ctDNA an in ST2/BN2 molecular subtype was a "powerful" biomarker that can determine which patients respond best to R-CHOP.

The trouble is that, without a proven intervention, it isn't clear what physicians should do with that information. "If you tell a patient that they're at more risk of progressing or dying from this disease, the next question from them will be OK, so what are you going to do about it?" Lakhotia said.

Such studies hold out hope for clinical use, but they technical challenges remain. "To have a unified assay that is applicable to a broader population, we need to identify 1 or 2 lead assays that can be studied more broadly across the lymphoma subtypes. The differences are primarily around the pre-analytical and the analytical part of the assay. The need to harmonize the processes more and hopefully have a commercially available assay, but that will take time," Lakhotia said.

REFERENCES

1. Larouche JF, Berger F, Chassagne-Clément C, et al. Lymphoma recurrence 5 years or later following diffuse large B-cell lymphoma: clinical characteristics and outcome. J Clin Oncol. 2010;28(12):2094-2100. doi:10.1200/JCO.2009.24.5860

2. Cheson BD, Fisher RI, Barrington SF, et al. Recommendations for initial evaluation, staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: the Lugano classification. J Clin Oncol. 2014;32(27):3059-3068. doi:10.1200/JCO.2013.54.8800

3. Greaves M, Maley CC. Clonal evolution in cancer. Nature. 2012;481(7381):306-313. doi:10.1038/nature10762

4. Kurtz DM, Esfahani MS, Scherer F, et al. Dynamic risk profiling using serial tumor biomarkers for personalized outcome prediction. Cell. 2019;178(3):699-713.e19. doi:10.1016/j.cell.2019.06.011.

5. Roschewski M, Rossi D, Kurtz DM, Alizadeh AA, Wilson WH. Circulating tumor DNA in lymphoma: principles and future directions. Blood Cancer Discov. 2022;3(1):5-15. doi:10.1158/2643-3230.BCD-21-0029

6. 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. 2015;16(5):541-549. doi:10.1016/S1470-2045(15)70106-3

7. Moia R, Talo. a D, Dondolin R, et al. Molecular clustering on ctDNA improves the prognostic stratification of DLBCL patients compared to ctDNA levels. Hematol Oncol. 2023;41(suppl 2):174-175. doi:10.1002/hon.3163_110