Tumor Heterogeneity in Diffuse Large B-cell Lymphoma in Relation to CNS Involvement and Cell-free DNA

Only about 60% of patients with diffuse large B-cell lymphoma (DLBCL) are cured with standard chemo- and immunotherapy, with R-CHOP(Rituximab, cyclophosphamide, doxorubicin hydrochloride, oncovin, prednisone).

Early and more accurate identification of patients with unfavorable prognosis after R-CHOP will enable the clinician to find patients who need other treatment strategies. New predictive biomarkers are needed to predict the effect of treatment at the individual level.

A proportion of patients have, or develop, lymphoma involvement of the central nervous system (CNS) and this group has a particularly dismal outcome. Identification of patients with primary subclinical or later manifest involvement of the CNS also requires new and improved diagnostic methods. Currently, a number of clinical parameters are included in a CNS-IPI risk score, that can be used to predict the risk of CNS disease. This score enables the clinicians to decide if patients should have prophylactic chemotherapy to prevent development of CNS disease. However, this risk model is far from accurate and the effect of prophylactic treatment is based on a low level of evidence.

DLBCL is a highly heterogeneous disease as evidenced by the 19 subtypes and variants of large B-cell lymphoma diagnoses found in the WHO classification of tumors of hematopoietic and lymphoid tissues. However, the cancer within each patient is also characterized by heterogeneity. The fact that different lymphoma cells may co-exist in the same tumor has been known for decades. One example of this is transformation of indolent lymphoma to a more aggressive lymphoma subtype, often DLBCL where the two types of lymphoma cells are found in the same lymph node. Differences between the cancer cells in the same tumor is termed intratumor heterogeneity. Such differences may be found within the same tumor as well as at spatially separate locations(inter-tumor heterogeneity). Intratumor heterogeneity may have an important impact on the clinical course and response to therapy. In clinical practice, the prognosis in patients with similar clinical stage and International Prognostic Index (IPI) score is often diverse. Such differences may be attributed to both intra and inter-tumor heterogeneity. Molecular profiling (mutations) in different nodal and extra nodal sites within the same patient have not before been studied in larger cohorts of DLBCL patients (Inter-tumor heterogeneity).

Monitoring circulating cell-free tumor DNA (ctDNA) has the potential to serve as a non-invasive tool for measuring the entire spectrum of genetic aberrations found in an individual patient as an alternative to more invasive diagnostic procedures.

Clinical presentation of DLBCL:

DLBCL is the most common type of lymphoma, and accounts for approximately 40% of non-Hodgkin lymphomas in adults in Denmark (500 new cases per year). DLBCL can arise de novo or from the histologic transformation of more indolent lymphomas, most commonly follicular lymphoma (FL) and chronic lymphocytic leukemia (CLL). DLBCL primarily affects elderly patients with a median age at diagnosis of nearly 70 years. Patients can present with both nodal and extra-nodal involvement. More than one extra-nodal involvement is a negative prognostic factor, and some sites are associated with higher risk of having or developing CNS involvement (e.g. testis, adrenal glands, uterus).

DLBCL is characterized as a clinically heterogeneous disease with a variable outcome, however it is potentially curable. R-CHOP is the standard therapy for patients with DLBCL. The 5-year progression-free survival (PFS) is around 60%. However, more than 30% of patients treated with R-CHOP do not respond or relapse, with the majority of these patients succumbing to their disease. Despite risk models like the International Prognostic Index (IPI) and CNS-IPI, patients with a high risk of treatment failure, relapse or CNS involvement cannot be identified accurately by standard prognostic factors. Therefore, new prognostic factors and biomarkers are under investigation at diagnosis, to better identify those high-risk patients, who may benefit from an alternative therapeutic strategy.

Project aims and perspectives:

Our current understanding of the molecular mechanisms associated with DLBCL relapse is limited. It is presently unclear whether clonal heterogeneity in primary tumors plays a role in DLBCL relapse. Our study is based on the hypothesis that divergent subclones give rise to the relapse tumor in an early-divergent scenario. This scenario suggest that several divergent clones exist at diagnosis besides the one that gives rise to relapse. Thus, information regarding intratumor heterogeneity and clonal evolution of the disease may provide a better understanding of DLBCL and development of prognostic factors regarding identification of high risk DLBCL patients. If the tumor is heterogenic, as studies of ctDNA have indicated, it could potentially transform the clinical care paradigm. It could perhaps explain the diverse outcome in patients with similarly clinical presentation, or why some patients develop CNS involvement. Furthermore, knowledge of tumor heterogeneity could guide the clinician in treatment choice and protocol development and thereby improve outcome for the patients.

If heterogeneity could be detected in ctDNA it could be made operational and the patients could avoid invasive biopsy procedures. The dynamic evolution of the tumor could perhaps also be monitored by ctDNA regarding tumor burden related to interim and end of treatment PET-CT and treatment intensity.

Current genomic models in DLBCL are mainly based on single tumor biopsies. They find a wide spectrum of mutations between patients. The single tumor models may underestimate the possible heterogeneity between tumor sites in the individual patient. Furthermore, heterogeneity, data on mutational evolution and relationship with CNS involvement largely remains unknown. Our setup will make it possible to clarify heterogeneity within the individual patient and investigate the relation to prognosis and CNS involvement.

Primary objective of the study:

• To identify the pattern and variations of mutations in different lymphoma sites in individual patients

Secondary objectives:

• To identify inter-tumor heterogeneity between nodal and extra nodal sites
• To identify the association between inter-tumor heterogeneity and risk of CNS involvement
• To identify clonal evolution and clonal relationship between primary tumor site and tumor at relapse
• To explore the role of "liquid biopsies" and ctDNA in monitoring inter-tumor heterogeneity
• To identify heterogeneity between nodal sites and bone-marrow biopsy.
• Inter-patient heterogeneity identified with our local clinical lymphoma panel can be used to subdivide patients according to the Wright classification

Clinical trials:

The project will consist of a retrospective study and an observational and clinical study.

5.1 Retrospective study - cohort A - Specific markers in patients with DLBCL

The first part of the retrospective study, cohort A, will be a study of different biopsies taken from the same patient. The biopsies from different nodal and/or extra nodal sites have either been taken at the same time or at different times in the treatment course. The time period is 2015-2019.

The aim with cohort A is to explore the biopsy samples for various purposes:

• The biopsies taken at different times: give us the possibility to look at clonal evolution over time.
• The biopsies, taken at the same time, but from different sites: gives us the possibility to look at inter-tumor heterogeneity in one patient.

The results from different sites from the same patient will be compared and used to subgroup the patient according to the Wright algorithm. All analyses will be performed on paraffin embedded biopsy material.

Retrospective study - cohort B - Specific markers in patients with DLBCL and secondary CNS involvement

In the retrospective study - cohort B, the investigators explore heterogeneity and clonal evolution between CNS lymphoma and nodal/extra-nodal sites using already collected biopsies.

The investigators will look at patients with both systemic disease and involvement of the central nervous system.

The time period is: 2010-2019

The aim with cohort B is to explore the biopsy samples for various purposes:

• The biopsies, taken from different sites (CNS vs systemic) gives us the possibility to look at inter-tumor heterogeneity and possible clonal evolution.
• The biopsies taken at different times: give us the possibility to look at clonal evolution over time

By comparing our results from the retrospective cohort A, the aim is to possibly define particularly important mutations that are only present within patients whose disease has spread to the central nervous system. Knowledge of mutations that may predispose to involvement of the CNS would be essential in identifying patients with this risk.

The results from different sites from the same patient will be compared and used to subgroup the patient according to the Wright algorithm. All analyses will be performed on paraffin embedded biopsy material.

Prospective study - cohort A and B - Heterogeneity within lymphoma patients

The prospective study will include two different cohorts:

A: DLBCL patients with lymphoma involvement of several nodal sites accessible for biopsy and bone-marrow.

B: DLBCL patients with both nodal, extra-nodal sites accessible for biopsy and bone-marrow.

The prospective study will include newly diagnosed patients with DLBCL. Patients will be included in either cohort A or B according to localization of the disease ascertained by PET-CT to clarify nodal vs extra-nodal involvement.

Cohort A - patients with more than one nodal site: One additional biopsy for clinical evaluation will be performed from a different location than where the diagnostic biopsy was performed. These biopsies can explore the heterogeneity within the patient at nodal sites.

Available bone-marrow biopsies will be analyzed as a separate site.

Cohort B - patients with both nodal and extra-nodal sites: One biopsy for clinical evaluation will be performed from a different site than the diagnostic biopsy, resulting in one nodal and one extra-nodal biopsy. These biopsies will explore the heterogeneity between nodal and extra-nodal sites within the patient.

Any available bone-marrow biopsies will be analyzed as a separate site.