Relapse Timing Is Associated With Distinct Evolutionary Dynamics in Diffuse Large B-Cell Lymphoma

PURPOSE

Diffuse large B-cell lymphoma (DLBCL) is cured in more than 60% of patients, but outcomes remain poor for patients experiencing disease progression or relapse (refractory or relapsed DLBCL [rrDLBCL]), particularly if these events occur early. Although previous studies examining cohorts of rrDLBCL have identified features that are enriched at relapse, few have directly compared serial biopsies to uncover biological and evolutionary dynamics driving rrDLBCL. Here, we sought to confirm the relationship between relapse timing and outcomes after second-line (immuno)chemotherapy and determine the evolutionary dynamics that underpin that relationship.

PATIENTS AND METHODS

Outcomes were examined in a population-based cohort of 221 patients with DLBCL who experienced progression/relapse after frontline treatment and were treated with second-line (immuno)chemotherapy with an intention-to-treat with autologous stem-cell transplantation (ASCT). Serial DLBCL biopsies from a partially overlapping cohort of 129 patients underwent molecular characterization, including whole-genome or whole-exome sequencing in 73 patients.

RESULTS

Outcomes to second-line therapy and ASCT are superior for late relapse (>2 years postdiagnosis) versus primary refractory (<9 months) or early relapse (9-24 months). Diagnostic and relapse biopsies were mostly concordant for cell-of-origin classification and genetics-based subgroup. Despite this concordance, the number of mutations exclusive to each biopsy increased with time since diagnosis, and late relapses shared few mutations with their diagnostic counterpart, demonstrating a branching evolution pattern. In patients with highly divergent tumors, many of the same genes acquired new mutations independently in each tumor, suggesting that the earliest mutations in a shared precursor cell constrain tumor evolution toward the same genetics-based subgroups at both diagnosis and relapse.

CONCLUSION

These results suggest that late relapses commonly represent genetically distinct and chemotherapy-naïve disease and have implications for optimal patient management.

Genetic subdivisions of follicular lymphoma defined by distinct coding and noncoding mutation patterns

Follicular lymphoma (FL) accounts for ∼20% of all new lymphoma cases. Increases in cytological grade are a feature of the clinical progression of this malignancy, and eventual histologic transformation (HT) to the aggressive diffuse large B-cell lymphoma (DLBCL) occurs in up to 15% of patients. Clinical or genetic features to predict the risk and timing of HT have not been described comprehensively. In this study, we analyzed whole-genome sequencing data from 423 patients to compare the protein coding and noncoding mutation landscapes of untransformed FL, transformed FL, and de novo DLBCL. This revealed 2 genetically distinct subgroups of FL, which we have named DLBCL-like (dFL) and constrained FL (cFL). Each subgroup has distinguishing mutational patterns, aberrant somatic hypermutation rates, and biological and clinical characteristics. We implemented a machine learning-derived classification approach to stratify patients with FL into cFL and dFL subgroups based on their genomic features. Using separate validation cohorts, we demonstrate that cFL status, whether assigned with this full classifier or a single-gene approximation, is associated with a reduced rate of HT. This implies distinct biological features of cFL that constrain its evolution, and we highlight the potential for this classification to predict HT from genetic features present at diagnosis.

Relapse timing is associated with distinct evolutionary dynamics in DLBCL

Diffuse large B-cell lymphoma (DLBCL) is cured in over 60% of patients, but outcomes are poor for patients with relapsed or refractory disease (rrDLBCL). Here, we performed whole genome/exome sequencing (WGS/WES) on tumors from 73 serially-biopsied patients with rrDLBCL. Based on the observation that outcomes to salvage therapy/autologous stem cell transplantation are related to time-to-relapse, we stratified patients into groups according to relapse timing to explore the relationship to genetic divergence and sensitivity to salvage immunochemotherapy. The degree of mutational divergence increased with time between biopsies, yet tumor pairs were mostly concordant for cell-of-origin, oncogene rearrangement status and genetics-based subgroup. In patients with highly divergent tumors, several genes acquired exclusive mutations independently in each tumor, which, along with concordance of genetics-based subgroups, suggests that the earliest mutations in a shared precursor cell constrain tumor evolution. These results suggest that late relapses commonly represent genetically distinct and chemotherapy-naïve disease.

Prognostic significance of FCGR2B expression for the response of DLBCL patients to rituximab or obinutuzumab treatment

Fc γ receptor IIB (FcγRIIB) is an inhibitory molecule capable of reducing antibody immunotherapy efficacy. We hypothesized its expression could confer resistance in patients with diffuse large B-cell lymphoma (DLBCL) treated with anti-CD20 monoclonal antibody (mAb) chemoimmunotherapy, with outcomes varying depending on mAb (rituximab [R]/obinutuzumab [G]) because of different mechanisms of action. We evaluated correlates between FCGR2B messenger RNA and/or FcγRIIB protein expression and outcomes in 3 de novo DLBCL discovery cohorts treated with R plus cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) reported by Arthur, Schmitz, and Reddy, and R-CHOP/G-CHOP-treated patients in the GOYA trial (NCT01287741). In the discovery cohorts, higher FCGR2B expression was associated with significantly shorter progression-free survival (PFS; Arthur: hazard ratio [HR], 1.09; 95% confidence interval [CI], 1.01-1.19; P = .0360; Schmitz: HR, 1.13; 95% CI, 1.02-1.26; P = .0243). Similar results were observed in GOYA with R-CHOP (HR, 1.26; 95% CI, 1.00-1.58; P = .0455), but not G-CHOP (HR, 0.91; 95% CI, 0.69-1.20; P = .50). A nonsignificant trend that high FCGR2B expression favored G-CHOP over R-CHOP was observed (HR, 0.67; 95% CI, 0.44-1.02; P = .0622); however, low FCGR2B expression favored R-CHOP (HR, 1.58; 95% CI, 1.00-2.50; P = .0503). In Arthur and GOYA, FCGR2B expression was associated with tumor FcγRIIB expression; correlating with shorter PFS for R-CHOP (HR, 2.17; 95% CI, 1.04-4.50; P = .0378), but not G-CHOP (HR, 1.37; 95% CI, 0.66-2.87; P = .3997). This effect was independent of established prognostic biomarkers. High FcγRIIB/FCGR2B expression has prognostic value in R-treated patients with DLBCL and may confer differential responsiveness to R-CHOP/G-CHOP.

Abstract PO-32: NFKBIZ 3′ UTR mutations confer selective growth advantage and affect drug response in diffuse large B-cell lymphoma

Introduction: The activated B cell-like (ABC) molecular subgroup of diffuse large B-cell lymphoma (DLBCL) is characterized by activation of NF-κB signaling. Recurrent mutations affecting genes such as MYD88, CD79A/B, and TNFAIP3 contribute to this in some cases, but there remain tumors with no known genetic basis for this pathway activation. This suggests that our understanding of ABC DLBCL drivers remains incomplete. Previously, NFKBIZ was shown to be amplified in 10% of ABC DLBCLs and to contribute to activation of NF-κB signaling. We recently described a novel pattern of mutations affecting the 3′ UTR of NFKBIZ resulting in an overall mutation rate of 30% (UTR or AMP) in ABC DLBCL. These NFKBIZ UTR mutations are mutually exclusive with MYD88 mutations, thus suggesting they may also lead to activation of NF-κB signaling. The NFKBIZ protein interacts with NF-κB transcription factors and is thought to regulate canonical NF-κB signaling. We hypothesized that NFKBIZ UTR mutations affect the normally rapid degradation of this mRNA by disrupting secondary structures recognized by RNA-binding proteins such as ribonucleases. The resulting elevated NFKBIZ mRNA levels would lead to accumulation of protein and may be a novel mechanism to promote cell growth and survival in ABC DLBCL.

Methods: NFKBIZ 3′ UTR mutations were introduced into a DLBCL cell line using the CRISPR-Cas9 system. A competitive growth assay with wild-type (WT) and CRISPR-mutant lines was performed to assess whether UTR mutations provide a growth advantage in culture (in vitro) and in mouse xenografts (in vivo). RNA-sequencing was then performed on WT and a subset of CRISPR-mutant lines and analyses were performed to identify genes upregulated by IκB-ζ in mutant lines. The IC50 of relevant drugs was determined by WST-1 assays after drug treatment on WT and mutant lines.

Results: Introduction of NFKBIZ mutations into a DLBCL cell line confirmed that UTR deletions lead to increased mRNA and protein levels. NFKBIZ UTR deletions give DLBCL cells a selective growth advantage over WT both in vitro and in vivo. RNA-sequencing of mutant and WT lines revealed possible transcriptional targets of NFKBIZ, including NF-κB targets and genes commonly overexpressed in ABC DLBCL. Novel candidate NFKBIZ targets were also discovered through this analysis, including CD274, the gene encoding PD-L1. Mutant cell lines had significantly higher IC50 compared to WT for the drugs ibrutinib, idelalisib, and masitinib, but not bortezomib, suggesting that NKFBIZ UTR mutations make cell lines more resistant to specific NF-κB pathway-targeted drugs.

Conclusions: This work directly establishes a role for NFKBIZ amplifications and 3′ UTR mutations in driving ABC DLBCL through NF-κB signaling. We demonstrate that these mutations can lead to overexpression of NFKBIZ and provide a selective growth advantage to cells both in vitro and in vivo. In addition, we found that these mutant lines were more resistant to some targeted lymphoma drugs but not others.

Citation Format: Sarah E. Arthur, Nicole Thomas, Jeffrey Tang, Christopher K. Rushton, Miguel Alcaide, Adèle Telenius, Shannon Healy, Anja Mottok, David W. Scott, Christian Steidl, Ryan D. Morin. NFKBIZ 3′ UTR mutations confer selective growth advantage and affect drug response in diffuse large B-cell lymphoma [abstract]. In: Proceedings of the AACR Virtual Meeting: Advances in Malignant Lymphoma; 2020 Aug 17-19. Philadelphia (PA): AACR; Blood Cancer Discov 2020;1(3_Suppl):Abstract nr PO-32.

Abstract IA42: Detecting and quantifying mutations associated with treatment resistance in aggressive lymphomas using ctDNA

A significant proportion of diffuse large B-cell lymphoma (DLBCL) patients treated with immunochemotherapy containing rituximab (R-CHOP) exhibit either primary or acquired treatment resistance. The advancement of therapeutics in the relapse setting has likely been encumbered by our limited understanding of the molecular features that underlie resistance to R-CHOP. Unfortunately, our knowledge of DLBCL genetics is mostly limited to analyses conducted on diagnostic tissue biopsies, which have not been exposed to the selective pressures imposed by therapy. Identifying genetic alterations that contribute to treatment resistance may reveal additional treatment options and lead to biomarkers allowing patients to be paired with appropriate treatments. Genetic subgroups are gaining popularity as a new strategy to implement precision medicine in DLBCL (1). The relevance of these and other biomarkers in the relapse setting remains unclear due to limited genetic exploration of relapsed and refractory DLBCL (rrDLBCL). Progress has been limited, in part, by the requirement of tissue biopsies collected after relapse. It is well established that quantitative genomic techniques such as digital PCR and targeted sequencing can be used to determine the proportion of tumor DNA in plasma from lymphoma patients (2). With a sufficiently broad panel, sequencing affords additional opportunities including the ability to identify subclonal structure and population dynamics over time. This presentation will discuss our recent analysis of a large collection of ctDNA primarily comprising DLBCL patients on various clinical trials (3). Targeted sequencing of these samples and comparison to exome data from a meta-cohort of previously characterized untreated DLBCL biopsies revealed six genes significantly enriched for mutations upon relapse. We found both TP53 and KMT2D were mutated in the majority of rrDLBCLs, and these mutations persisted in the dominant clone following relapse, suggesting a role in primary treatment resistance. By inferring subclonal dynamics, we observed recurrent patterns of clonal expansion and contraction following rituximab-based therapy, with MS4A1 mutations representing the only example of consistent clonal expansion. MS4A1 missense mutations within the transmembrane domains led to loss of CD20 expression in vitro, and patient tumors harboring these mutations lacked CD20 protein expression. Our analysis nominates TP53 and KMT2D mutation status as novel prognostic factors that may facilitate the identification of high-risk patients prior to therapy. Moreover, we have demonstrated the potential to identify tumors with loss of CD20 surface expression stemming from MS4A1 mutations. Implementation of noninvasive assays to detect such features of acquired treatment resistance may allow timely transition to more effective treatment regimens. In certain scenarios whole-exome sequencing (WES) or whole-genome sequencing (WGS) can be successfully applied to ctDNA, thereby allowing the identification of mutations, structural variation, and copy number changes. Low-pass sequencing of shotgun libraries can also be used to ascertain course estimates of ctDNA levels as well as the copy number landscape (4). Given the importance of copy number and structural alterations in the inference of genetic subgroups, these methods may allow the exploration of these groups and their stability over time. Through a series of illustrative examples, this presentation will explore the benefits of each of these techniques in the study of tumor evolution and acquired treatment resistance in DLBCL.

References: 1. Morin RD, Scott DW. DLBCL subclassification: Divide and conquer? Blood 2020;135:1722–4. 2. Rossi D et al. The development of liquid biopsy for research and clinical practice in lymphomas: Report of the 15-ICML workshop on ctDNA. Hematol Oncol 2020;38:34–7. 3. Rushton CK et al. Genetic and evolutionary patterns of treatment resistance in relapsed B-cell lymphoma. Blood Adv 2020;4:2886–98. 4. Adalsteinsson VA et al. Scalable whole-exome sequencing of cell-free DNA reveals high concordance with metastatic tumors. Nat Commun 2017;8:1324.

Citation Format: Nicole Thomas, Laura K. Hilton, Neil Michaud, Kevin Bushell, Ryan Rys, Michael Jain, Lois Shepherd, Marco A. Marra, John Kuruvilla, Michael Crump, Koren Mann, Sarit Assouline, Christian Steidl, Mark S. Cragg, David W. Scott, Nathalie Johnson, Ryan D. Morin, Christopher K. Rushton, Sarah E. Arthur, Miguel Alcaide, Matthew Cheung, Aixiang Jiang, Krysta M. Coyle, Kirstie L. S. Cleary. Detecting and quantifying mutations associated with treatment resistance in aggressive lymphomas using ctDNA [abstract]. In: Proceedings of the AACR Virtual Meeting: Advances in Malignant Lymphoma; 2020 Aug 17-19. Philadelphia (PA): AACR; Blood Cancer Discov 2020;1(3_Suppl):Abstract nr IA42.

PRPS-ST: A protocol-agnostic self-training method for gene expression-based classification of blood cancers

Gene expression classifiers are gaining increasing popularity for stratifying tumors into subgroups with distinct biological features. A fundamental limitation shared by current classifiers is the requirement for comparable training and testing data sets. Here, we describe a self-training implementation of our probability ratio-based classification prediction score method (PRPS-ST), which facilitates the porting of existing classification models to other gene expression data sets. In comparison to gold standards, we demonstrate favorable performance of PRPS-ST in gene expression-based classification of DLBCL and B-ALL using a diverse variety of gene expression data types and pre-processing methods, including in classifications with a high degree of class imbalance. Tumors classified by our method were significantly enriched for prototypical genetic features of their respective subgroups. Interestingly, this included cases that were unclassifiable by established methods, implying the potential enhanced sensitivity of PRPS-ST.

Genetic and evolutionary patterns of treatment resistance in relapsed B-cell lymphoma

Diffuse large B-cell lymphoma (DLBCL) patients are typically treated with immunochemotherapy containing rituximab (rituximab, cyclophosphamide, hydroxydaunorubicin-vincristine (Oncovin), and prednisone [R-CHOP]); however, prognosis is extremely poor if R-CHOP fails. To identify genetic mechanisms contributing to primary or acquired R-CHOP resistance, we performed target-panel sequencing of 135 relapsed/refractory DLBCLs (rrDLBCLs), primarily comprising circulating tumor DNA from patients on clinical trials. Comparison with a metacohort of 1670 diagnostic DLBCLs identified 6 genes significantly enriched for mutations upon relapse. TP53 and KMT2D were mutated in the majority of rrDLBCLs, and these mutations remained clonally persistent throughout treatment in paired diagnostic-relapse samples, suggesting a role in primary treatment resistance. Nonsense and missense mutations affecting MS4A1, which encodes CD20, are exceedingly rare in diagnostic samples but show recurrent patterns of clonal expansion following rituximab-based therapy. MS4A1 missense mutations within the transmembrane domains lead to loss of CD20 in vitro, and patient tumors harboring these mutations lacked CD20 protein expression. In a time series from a patient treated with multiple rounds of therapy, tumor heterogeneity and minor MS4A1-harboring subclones contributed to rapid disease recurrence, with MS4A1 mutations as founding events for these subclones. TP53 and KMT2D mutation status, in combination with other prognostic factors, may be used to identify high-risk patients prior to R-CHOP for posttreatment monitoring. Using liquid biopsies, we show the potential to identify tumors with loss of CD20 surface expression stemming from MS4A1 mutations. Implementation of noninvasive assays to detect such features of acquired treatment resistance may allow timely transition to more effective treatment regimens.

Genome-wide discovery of somatic coding and noncoding mutations in pediatric endemic and sporadic Burkitt lymphoma

Although generally curable with intensive chemotherapy in resource-rich settings, Burkitt lymphoma (BL) remains a deadly disease in older patients and in sub-Saharan Africa. Epstein-Barr virus (EBV) positivity is a feature in more than 90% of cases in malaria-endemic regions, and up to 30% elsewhere. However, the molecular features of BL have not been comprehensively evaluated when taking into account tumor EBV status or geographic origin. Through an integrative analysis of whole-genome and transcriptome data, we show a striking genome-wide increase in aberrant somatic hypermutation in EBV-positive tumors, supporting a link between EBV and activation-induced cytidine deaminase (AICDA) activity. In addition to identifying novel candidate BL genes such as SIN3A, USP7, and CHD8, we demonstrate that EBV-positive tumors had significantly fewer driver mutations, especially among genes with roles in apoptosis. We also found immunoglobulin variable region genes that were disproportionally used to encode clonal B-cell receptors (BCRs) in the tumors. These include IGHV4-34, known to produce autoreactive antibodies, and IGKV3-20, a feature described in other B-cell malignancies but not yet in BL. Our results suggest that tumor EBV status defines a specific BL phenotype irrespective of geographic origin, with particular molecular properties and distinct pathogenic mechanisms. The novel mutation patterns identified here imply rational use of DNA-damaging chemotherapy in some patients with BL and targeted agents such as the CDK4/6 inhibitor palbociclib in others, whereas the importance of BCR signaling in BL strengthens the potential benefit of inhibitors for PI3K, Syk, and Src family kinases among these patients.