Primary Mediastinal B-cell Lymphoma Patients Research Articles

Omission of Radiotherapy in Primary Mediastinal B-Cell Lymphoma: IELSG37 Trial Results.

The role of consolidation radiotherapy in primary mediastinal B-cell lymphoma (PMBCL) patients is controversial. The IELSG37 trial, a randomized non-inferiority study, aimed to assess whether irradiation can be omitted in PMBCL patients with complete metabolic response (CMR) after induction immunochemotherapy. Primary endpoint was progression-free survival (PFS) at 30 months post-randomization. Patients with CMR were randomly assigned to observation or consolidation radiotherapy (30 Gy). With a non-inferiority margin of 10% (assuming a 30-month PFS of 85% in both arms), a sample size of 540 patients was planned with 376 expected to be randomized. The observed events were considerably lower than expected, therefore, primary endpoint analysis was conducted when ≥95% of patients were followed for ≥30 months. Of 545 patients enrolled, 268 were in CMR after induction and were randomized to observation (n=132) or radiotherapy (n=136). The 30-month PFS was 96.2% in the observation arm and 98.5% in the radiotherapy arm, with a stratified hazard ratio of 1.47 (95%CI, 0.34 to 6.28) and absolute risk difference of 0.68% (95%CI, -0.97% to 7.46%). The 5-year overall survival was 99% in both arms.Non-randomized patients were managed according to local policies. Radiotherapy was the only treatment in 86% of those with Deauville score (DS) 4 and in 57% of those with DS 5. The 5-year PFS and OS of patients with DS 4 (95.8% and 97.5%, respectively) were not significantly different from those of randomized patients. Patients with DS5 had significantly poorer 5-year PFS and OS (60.3% and 74.6%, respectively). This study, the largest randomized trial of radiotherapy in PMBCL, demonstrated favorable outcomes in patients achieving CMR with no survival impairment for those omitting irradiation.

Management and Outcomes for Primary Mediastinal B-Cell Lymphoma Patients with Partial Metabolic Response

Background: The optimal management of patients with primary mediastinal B-cell lymphoma (PMBCL) who experience a partial metabolic response (PMR) on FDG-PET after initial systemic therapy is unclear. This study aimed to investigate different management strategies and outcomes for PMBCL patients who achieved PMR following primary systemic therapy, utilizing a single-center retrospective design. Methods: We reviewed the electronic medical records of 121 PMBCL patients who underwent post-chemotherapy PET scans at the Princess Margaret Cancer Center, Toronto, from January 2009 to September 2021. Using Modified Lugano criteria (2014), post-chemotherapy PET scan results were evaluated, with Deauville score (DS) 0-3 considered a complete metabolic response. Demographic, clinical, and imaging characteristics of the PMBCL population were analyzed. The Kaplan-Meier method was used to estimate progression-free survival (PFS), measured from the time of the post-systemic therapy PET. Results: A total of 121 patients with PMBCL underwent post-chemotherapy PET scans. Among them, 49 patients (40%) demonstrated incomplete metabolic response (DS4-5) Among these 49 patients, 12 showed primary refractory disease, while 37 demonstrated a partial metabolic response (PMR, DS4). The median age at diagnosis of the 37 patients with PMR was 30.8 years (range:18.4 - 52.8 years), and 20 (54%) were female. The median length of follow-up for these patients was 3.7 years (range: 0.4 to 9.2 years). Most PMR patients were stage I/II (26 patients, 70.3%), while the remaining were stage III (2 patients, 5.4%), or stage IV (8 patients, 21.6%), and 26 (70.3%) had a large mediastinal mass (maximum diameter exceeding 10cm). Initial systemic treatment was R-CHOP-21 (34 patients, 91.9%), R-CHOP combined with dose-adjusted R-EPOCH (2 patients, 5.4%), and dose-adjusted EPOCH (1 patient, 2.7%). Thirty one patients (83.8%) received 6 cycles. Following treatment strategies were employed to manage PMR: 1) radiation therapy (RT) only with or without biopsy in 30 patients (80.1%); 2) an observational approach (repeating PET scans without immediate treatment) in 3 patients (8.1%); 3) salvage chemotherapy in 3 patients (8.1%) and other approach in 1(2.7%) patient. The 2-year progression-free survival (PFS) rate for all patients with incomplete metabolic response (n=49) was 74.5% (95% CI = 62.5% to 88.9%). For patients with primary progression on post-chemotherapy PET scans the 2-year PFS (ie free of 2nd progression) was 40% (18.7% to 85.5%), while the 2-year PFS for patients with PMR was 85.4% (74.3% to 98.1%). PMR patients who underwent radiotherapy (RT) had the 2-year PFS rate of 92.7% (95% CI: 83.5% to 100.0%). No progression or relapsed occurred among the 3 patients with PMR who were managed with repeating PET scans without immediate further treatment. Conclusion: This study replicates prior work demonstrating a higher rate of incomplete metabolic response for PMBCL patients following R-CHOP-21 than typically reported for DLBCL. RT to persistent PET-avid sites is associated with excellent PFS. The favorable outcome of PMR patients without any additional treatment illustrates the non-trivial risk of false-positive PET scans following systemic therapy, and the need to identify criteria that can distinguish those who need additional treatment from those who can be safely observed.

Cooperative Somatic Alterations in XPO1 and TNFAIP3 (A20) Promote Immune Modulation and Tissue Infiltration in Primary Mediastinal B Cell Lymphoma

Background: Primary Mediastinal B cell lymphoma (PMBCL) is an aggressive large B cell non-Hodgkin lymphoma typically presenting as a mass in the anterior mediastinum in early adulthood, disproportionately affecting females (2:1). Previous studies have identified constitutive activation of the NF-κB pathway and immune evasion as hallmark features of PMBCL. The master negative regulator of canonical NF-κB, TNFAIP3 (A20), is inactivated in ~25-60% of PMBCL cases via disruptive mutations or deletions. Gains of the 2p16.1-15 locus ( REL, XPO1, BCL11A) are seen in ~40-75% of PMBCL cases. Interestingly, PMBCL cases with 2p16.1-15 gain/amplification display concomitant increases in XPO1 expression but lack correlative increases in REL or BCL11A. In addition, XPO1 mutations are seen in ~15-33% of PMBCL cases, with the most prevalent mutation (E571K) implicated in NF-KB activation. Despite the established occurrence of genetic events impacting XPO1 and A20 in PMBCL, the cooperative potential of these alterations remains unexplored. Consequently, we conducted an extensive review of literature describing the PMBCL genomic landscape and utilized in vivo models to examine the combined potential of XPO1 and A20 alterations to promote lymphomagenesis. Methods: Genomic data was obtained from NGS analyses from a total of 233 PMBCL patients across 6 independent published research studies and analyzed for cooccurrence of A20 & XPO1 alterations (Camus et al. 2022; Chapuy et al. 2019; Lacy et al. 2020; Mareschal et al. 2015; Mottock et al. 2019; Tuveri et al 2022). For in vivo analysis, we developed a murine model of B cell specific overexpression of wildtype (WT) or E571K- XPO1 with CD19.cre driven depletion of A20 (Eµ-XPO1xA20 KO). Whole body histopathologic evaluation of FFPE tissue with H&E staining was performed on 6-month-old, age and sex matched littermate mice. Results: We found co-occurrence of inactivating events in A20 and XPO1 missense mutations impacting 15% of PMBCL evaluated cases, with consistent results seen across all studies (range: 13.8-20%). As expected, A20 mutations/del were more prevalent than XPO1 mutations across all studies (OR 2.46, 95% CI 0.84-7.18, p = 0.02). XPO1 mutated cases trended towards a greater likelihood of having co-occurring A20 mutations (OR 8.51, 95% CI 0.84-86.47, p = 0.08). XPO1 mutations and gain (2p16) were found to be mutually exclusive events with gains co-occurring with A20 mutations/del in ~17% of PMBCL cases evaluated. Analysis of survival of PMBCL patients with cooccurring A20 inactivating alterations and XPO1 mutations is ongoing although preliminarily trend towards reduced mean 2-year survival (p=0.056) compared to all others are observed. Taken together, co-occurring genetic alterations effecting A20 and XPO1 are prevalent in PMBCL the alternative biology stemming from potential cooperative mechanisms may represent an aggressive subgroup of PMBCL. To evaluate concurrent B cell-specific alterations of A20 and XPO1 in vivo we utilized Eµ-XPO1xA20 KO mouse models. Performing histopathology in 6-month-old mice we observed enlarged lymph nodes due to marked increases in lymphocytes within the cortex, paracortex, medulla, and sinuses in mice with loss of functional A20, particularly in Eµ-XPO1xA20 KO mice with overexpression of WT XPO1. Increased lymphocyte populations drove overall enlargement of the node (mandibular, mesenteric, pancreatic, and mediastinal) and distorted normal architecture, featuring small lymphocytes populating sinus regions and larger lymphocytes populating areas resembling follicles. Eµ-XPO1(WT & E571K)xA20 KO mice additionally featured mild mononuclear infiltrates in the lung and salivary glands; all of which were observed with increased severity compared to A20KO, Eµ-XPO1, or non-transgenic littermates. Significance: We identified co-occurring genetic lesions impacting A20 and XPO1 in PMBCL tumors and show that these alterations synergize in vivo leading to nodal/extranodal lymphocyte infiltration. The immunomodulatory effects appear to be driven by the cooperation of XPO1 overexpression, more so than mutation, and A20 inactivation in young mice. Further investigation of these models will allow elucidation of underlying cooperative mechanisms and reveal targetable pathways in PMBCL.

Anti-PD-1 Antibody (Tislelizumab) Combined with R-CHOP for the Treatment of Previously Untreated Primary Mediastinal B-Cell Lymphoma: A Prospective Phase II Study

Background: Primary mediastinal B-cell lymphoma (PMBCL) is a rare aggressive B-cell lymphoma of thymic origin. Due to the low incidence rate of PMBCL, there is currently no internationally accepted standard first-line treatment regimen. Dose-adjusted (DA) EPOCH-R which is the more commonly used first-line therapy plan is not so effective because patients usually cannot tolerate it. The R-CHOP regimen is comparable to the R-DA-EPOCH for the first-line treatment of PMBCL in terms of survival and has a better safety profile (Bhatt, V.R. et al, Cancer Treat Rev 2015; Iwanicka el al., Curr Hematol Malig Rep 2014). Meanwhile, patients with PMBCL usually have alterations of 9p24.1, leading to overexpression of PDL-1/2. Previous studies have found that PD-1 monoclonal antibody shows good efficacy in recurrent and refractory PMBCL (Steidl et al., Blood 2011; Zinzani et al., Blood 2017). Tislelizumab, a humanized IgG4 antibody that binds PD-1 with high affinity, minimizes binding to Fcγ receptor 1 (FcγRI) on macrophages by the genetic engineering of Fc segment, abrogating antibody-dependent phagocytosis (ADCP)-induced T-cell clearance. The efficacy and safety of its combination with R-CHOP in previously untreated PMBCL is not yet clear. We speculate that the combination of both may be a new option for the first-line therapy for PMBCL patients with good efficacy and low toxicity. Methods: This is a single-arm, multi-center, phase II clinical study (ChiCTR2100044313). Patients aged 18 years or older with a diagnosis of PMBCL and previously untreated were eligible. Eligible patients received 6 cycles of Tislelizumab (200 mg D1), Rituximab (375 mg/m 2 D1), Cyclophosphamide (750 mg/m 2 D2), Pirarubicin (50 mg/m 2 D2), Vincristine (1.4 mg/m 2, maximum dose 2 mg, D2), Prednisone (100 mg/d D2-6), every 21 days for 1 cycle. Disease response assessments were performed by positron emission tomography and computed tomography (PET-CT) or enhanced CT every two cycles, meanwhile peripheral blood ctDNA was monitored, and treatment-related adverse effects were assessed and recorded. The primary endpoint was objective remission rate (ORR). The secondary endpoints included treatment safety, progression-free survival (PFS) and overall survival (OS), and the exploratory analyses of the correlation between PD-L1 expression in tumor tissues, tumor-associated macrophages (TAM) phenotype in tumor microenvironment and efficacy. Results: From June 2020 to June 2023, 12 patients were enrolled and 9 patients had evaluable efficacy with the median age of 34 years, 3 males and 6 females, all patients had large mediastinal mass at initial diagnosis. 7 patients had PD-L1 expression (VENTANA SP263) above 80% and 2 cases could not be further tested due to insufficient tissue specimens. The ORR was 100%, and the CR rate was 77.8% (7/9) after 6 cycles of treatment. The 5 th and 9 th patient reached PR after 6 cycles of treatment due to irregular hospitalization. And CR was achieved after 2 cycles of regular treatment for the 5 th patient. NGS from formalin-fixed paraffin-embedded (FFPE) and circulating tumor DNA (ctDNA) evaluation from blood were performed across all patients before treatment. 5 patients underwent end of treatment-ctDNA test. Median follow-up 15m (6.8m-26.3m), the third patient was lost to follow-up at 9m. The other patients who achieved CR are still at complete remission. Moreover, we found SOCS1, TNFAIP3 are the most frequently mutated genes in biopsies of PMBCL and only the 5th patient had RHOA mutation which is highly related to PMBCL. The PET-CT and ctDNA assessments were negative in three patients at the end of 6 cycles with 60% concordance, and the 5 th and 9 th patient achieved PR on PET-CT, but the ctDNA assessments were negative. The most common adverse events (AEs) were gastrointestinal reactions, 2 patients experienced grade IV myelosuppression and 1 patient had grade I hyperthyroidism. No treatment-related deaths occurred. Conclusion: Tislelizumab combined with R-CHOP for the treatment of previously untreated PMBCL had good efficacy and sustained response with low toxicity. Longer follow-up is required to access the survival benefit of this regimen. Figure 1: Disease Remission in the Included Patients

Exploring the cell-free total RNA transcriptome in diffuse large B-cell lymphoma and primary mediastinal B-cell lymphoma patients as biomarker source in blood plasma liquid biopsies.

Diffuse large B-cell lymphoma (DLBCL) and primary mediastinal B-cell lymphoma (PMBCL) are aggressive histological subtypes of non-Hodgkin's lymphoma. Improved understanding of the underlying molecular pathogenesis has led to new classification and risk stratification tools, including the development of cell-free biomarkers through liquid biopsies. The goal of this study was to investigate cell-free RNA (cfRNA) biomarkers in DLBCL and PMBCL patients. Blood plasma samples (n=168) and matched diagnostic formalin-fixed paraffin-embedded (FFPE) tissue samples (n=69) of DLBCL patients, PMBCL patients and healthy controls were collected between 2016-2021. Plasma samples were collected at diagnosis, at interim evaluation, after treatment, and in case of refractory or relapsed disease. RNA was extracted from 200 µl plasma using the miRNeasy serum/plasma kit and from FFPE tissue using the miRNeasy FFPE kit. RNA was subsequently sequenced on a NovaSeq 6000 instrument using the SMARTer Stranded Total RNA-seq pico v3 library preparation kit. Higher cfRNA concentrations were demonstrated in lymphoma patients compared to healthy controls. A large number of differentially abundant genes were identified between the cell-free transcriptomes of DLBCL patients, PMBCL patients, and healthy controls. Overlap analyses with matched FFPE samples showed that blood plasma has a unique transcriptomic profile that significantly differs from that of the tumor tissue. As a good concordance between tissue-derived gene expression and the immunohistochemistry Hans algorithm for cell-of-origin (COO) classification was demonstrated in the FFPE samples, but not in the plasma samples, a 64-gene cfRNA classifier was developed that can accurately determine COO in plasma. High plasma levels of a 9-gene signature (BECN1, PRKCB, COPA, TSC22D3, MAP2K3, UQCRHL, PTMAP4, EHD1, NAP1L1 pseudogene) and a 5-gene signature (FTH1P7, PTMAP4, ATF4, FTH1P8, ARMC7) were significantly associated with inferior progression-free and overall survival in DLBCL patients, respectively, independent of the NCCN-IPI score. Total RNA sequencing of blood plasma samples allows the analysis of the cell-free transcriptome in DLBCL and PMBCL patients and demonstrates its unexplored potential in identifying diagnostic, cell-of-origin, and prognostic cfRNA biomarkers.

Frequent ZNF217 mutations lead to transcriptional deregulation of interferon signal transduction via altered chromatin accessibility in B cell lymphoma

Recent exome-wide studies discovered frequent somatic mutations in the epigenetic modifier ZNF217 in primary mediastinal B cell lymphoma (PMBCL) and related disorders. As functional consequences of ZNF217 alterations remain unknown, we comprehensively evaluated their impact in PMBCL. Targeted sequencing identified genetic lesions affecting ZNF217 in 33% of 157 PMBCL patients. Subsequent gene expression profiling (n = 120) revealed changes in cytokine and interferon signal transduction in ZNF217-aberrant PMBCL cases. In vitro, knockout of ZNF217 led to changes in chromatin accessibility interfering with binding motifs for crucial lymphoma-associated transcription factors. This led to disturbed expression of interferon-responsive and inflammation-associated genes, altered cell behavior, and aberrant differentiation. Mass spectrometry demonstrates that ZNF217 acts within a histone modifier complex containing LSD1, CoREST and HDAC and interferes with H3K4 methylation and H3K27 acetylation. Concluding, our data suggest non-catalytic activity of ZNF217, which directs histone modifier complex function and controls B cell differentiation-associated patterns of chromatin structure.

Observation vs. radiotherapy in primary mediastinal B-cell lymphoma patients with complete response to standard immunochemotherapy: The IELSG37 randomized trial.

LBA7505 Background: Primary mediastinal B-cell lymphoma (PMBCL) has a good prognosis if remission is rapidly achieved with dose‐intensive immunochemotherapy. In these patients mediastinal radiotherapy (RT) may consolidate responses; however, it increases the risk of second malignancies and coronary or valvular heart disease. The IELSG37 trial was planned with a non-inferiority design to test whether RT can be omitted in patients who achieve a complete metabolic response (CMR) after immunochemotherapy. Methods: Patients with newly diagnosed PMBCL were eligible. Initial rituximab and anthracycline-based therapy was chosen according to local practice. CMR was defined, upon central review of positron emission computed tomography (PET/CT) scans, as Deauville score 1 to 3, according to the Lugano classification. Responding patients were randomized to observation (OBS) or consolidation RT (30 Gy). Randomization was stratified on gender, chemotherapy regimen, country, and PET/CT score. The primary endpoint was progression-free survival (PFS) after randomization. The sample size (540 patients to enrol, and 376 to randomize) was calculated assuming a 30-month PFS probability of 0.85 in both arms, with alpha at 0.05, 80% power and a hazard ratio (HR) of 1.77 as non-inferiority margin. Results: At a median follow-up of 30 months, the number of observed events was considerably lower than expected, hence, the Independent Data Monitoring Committee (IDMC) of the trial recommended to complete the planned total accrual without increasing the study size or duration. The primary endpoint analysis was performed, according to the IDMC recommendation, with ≥80% of patients having a minimum follow-up of 30 months. 545 patients (209 men, 336 women) were enrolled. Induction immunochemotherapy was completed and response assessed in 530 patients, 268 of them (50.6%) achieved a CMR and were randomly allocated to OBS (n = 132) or RT (n = 136). The PFS at 30 months was 98.5% (95%CI, 94.3 – 99.6) in the RT arm and 96.2% (95%CI, 91.1-98.4) in the OBS arm (P = 0.278). The estimated relative effect of radiotherapy vs observation in terms of hazard ratio (HR) was 0.47 (0.12-1.89) without adjustments and 0.79 (0.19-3.31) after stratification for the variables used for randomization. At 30 months the absolute risk reduction from RT was 2.3% (-1.5 to 6.2) unadjusted, and 0.8% (-3.0 to 8.3) with stratified HR. The number needed to treat is high (43 patients, unadjusted, and 126 after stratification). The 5-year overall survival was 99% in both arms. Longer follow-up is needed to examine late toxicity. Conclusions: This study is the largest prospective study of PMBCL ever conducted and although the event rate did not reach the assumed level, its evidence supports the omission of RT in patients achieving a CMR after immunochemotherapy.

Abstract 1913: Population-based analysis of primary mediastinal B-cell lymphoma: a look at Hispanic outcomes

Abstract Background: Primary Mediastinal B-cell Lymphoma (PMBL) is a rare, aggressive Non-Hodgkin Lymphoma (NHL) accounting for approximately 2-5% of all NHL (PMID: 34330673). Due to its rarity, there is scarce literature on the demographic features of PMBL. Moreover, there is little to no research on outcomes of ethnic minority groups, particularly in Hispanics (HI) (PMID: 27399089.) Thus, there is a need for further investigation of disease characteristics to advance treatment options for the future. This is a national scale population-based analysis examining disparities in HI vs Non-Hispanics (NH) and its impact on survival outcomes in patients with PMBL. Material and Methods: Data was analyzed on PBML patients in the United States reported to the Surveillance, Epidemiology, and End Results (SEER) database between 2000 and 2018. SEER contains the most comprehensive population-based cancer information in the U.S., covering approximately 27% of the total US population, and up to 36% of Hispanics alone. Racial groups analyzed included non-Hispanic whites, Hispanic whites, blacks, and Asians/PIs (Pacific Islanders). Patient characteristics, age-adjusted incidence rate, and survival rate were compared across ethnic groups, Hispanics (HI) vs Non-Hispanics (NH). Stratification by age, gender, and stage at diagnosis was considered. Kaplan-Meier and Cox regression analyses were used to compare overall survival (OS) between HI and NH. Multivariate analysis and propensity score matching were performed with adjustment for age, stage and B-symptoms. Results: Of 1,014 PMBL patients, 15% were HI, and 85% were NH. PMBL affects mostly women in both HI and NH (66% and 60% p=0.215). HI were diagnosed at a younger median age, 30 y.o vs 36 y.o (p=0.001), compared to NH. The majority of HI and NH were diagnosed under 40 years old (p=0.010). Both groups were mostly diagnosed between 2015-2018 (44% vs 41%) (p=0.768). The majority of both groups did not receive radiation, however it was noted that HI received less radiation compared to NH (72% vs 66%) (p=0.180). The median survival time was not reached indicating a favorable prognosis. On survival analysis; the survival probability of HI vs NH at 2 years was 0.87 vs 0.86, at 5 years 0.85 vs 0.85, and 10 years 0.83 vs 0.82. and there was not an OS difference favoring HI/NH (p = 0.94). On multivariate analysis, when adjusting for age, those who were older than 80 y.o and between 60 to 80 y.o, had worse OS compared to those younger than 60 y.o with HR 9.3 (95% CI: 4.3 - 20) and 8.8 (95% CI: 4 - 19). Conclusions: Our study analysis shows that the demographic variables between HI and NH were generally homogenous. In patients with PMBL, HI were diagnosed younger compared to their NH counterparts, however this did not affect OS. In conclusion, HI and NH patients with PMBL had similar treatment and outcomes showing that, when there is equity in health, favorable prognosis is possible with the current standard of care. Citation Format: Lauren Diaz Boyle, Esteban Toro-Velez, Adolfo Enrique Diaz Duque. Population-based analysis of primary mediastinal B-cell lymphoma: a look at Hispanic outcomes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 1913.

Improved Survival Outcomes for Intensive Versus Standard Chemoimmunotherapy in Primary Mediastinal B-Cell Lymphoma: A Meta-Analysis of 4068 Patients

Background Primary mediastinal B-cell lymphoma (PMBCL) is a distinct clinicopathologic entity with a predilection for female adolescents and young adults (AYAs). Due to its rarity, there is a paucity of prospective data and lack of randomized studies to inform on optimal frontline therapy. Expert guidelines from around the globe vary considerably with recommendations ranging from dose-intensive chemoimmunotherapy (CIT) to R-CHOP-21 with consolidative mediastinal radiation (RT). To assess if there are distinct differences in outcome between these approaches, we performed a systematic review and meta-analysis comparing dose-intensive versus standard CIT for the frontline treatment of PMBCL. Methods We performed a comprehensive and specific literature search on the frontline treatment of PMBCL. Studies included in our meta-analysis were prospective or retrospective published datasets that reported treatment outcomes (event free survival, EFS; progression free survival, PFS; overall survival, OS) for specific CIT regimens for PMBCL patients of any age. Standard approach was defined as R-CHOP-21 or CHOP-21, with or without RT. Dose intensive approaches were defined as regimens that increase the frequency, dose or number of systemic agents in comparison to R-CHOP-21. Case reports, small case series (<5 PMBCL patients) and unpublished conference abstracts were excluded. We searched MEDLINE, Embase, and Cochrane CENTRAL via Ovid and Web of science for all published literature on this topic on February 8th, 2022. This study followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline and the PRISMA extension statement.1 Two independent authors screened all studies to compile a final list of included publications, from which data was extracted. Patient number, trial type, treatment, consolidative RT/autologous stem cell transplant (auto-SCT), EFS, PFS and OS data was collected from each study. Pooled analysis for each efficacy outcome was calculated using percentages and sample size. Linear regression was used to compare treatment modalities for EFS, PFS and OS outcomes. Median follow-up time varied for each study, therefore outcomes were standardized by calculating average yearly event-free percentages. Results Our systematic review generated 1404 articles, which led to the inclusion of 53 publications, encompassing 4068 patients. 2517 patients were treated with dose-intensive approaches compared to 1551 with standard R-CHOP-21 (n=1095) or CHOP-21. Dose intensive approaches are summarized in Table 1. Consolidative RT was administered in 16.9% of the dose intensive arms, compared to 51.3% of R-CHOP regimens. Median follow up across all studies was 56 months. Outcome data is summarized in Table 1. 2234 patients in the dose intensive studies were evaluable for OS, with a cumulative 87.1% reported survival. Standard CIT studies reported 1440 patients with OS data, with a combined 78.9% reported survival. Controlling for differences in median follow up, the average percent per year surviving in the dose intensive CIT arm was 23.2% (95% CI 22.5 - 23.9) compared to 20.4% (95% CI 19.6 - 21.2) for standard CIT (p < 0.001). Aggregate overall survival and percent per year surviving for DA-EPOCH-R patients alone was 91.8% and 31.0% (29.6 - 32.3), compared to 87.3% and 24.8% (23.8 - 25.9; p < 0.001) for R-CHOP-21. Collective data revealed an EFS of 76.1% (n=794) and a PFS of 79.8% (n=1501) for the dose intensive group, compared to 55.7% (n=248) and 70.4% (n=1206) for the standard CIT arm. Improved EFS and PFS in the dose-intensive arm was confirmed with average percent per year event free survival (EFS 16.3% vs. 10.3%, p < 0.001; PFS 20.8% vs 18.7%, p < 0.001) for all patients, as well as the subgroup of DA-EPOCH-R compared to R-CHOP-21 (EFS 21.0% vs 15.6%, p < 0.001; PFS 35.4% vs 18.0%, p < 0.001). Conclusion To our knowledge, this is the largest meta-analysis reported to date of efficacy outcomes for frontline CIT in PMBCL. Pooled and yearly event-free analyses demonstrate that dose intensive approaches have more favorable overall survival, EFS and PFS. Furthermore, fewer patients in the dose intensive arm are exposed to consolidative RT, curtailing chronic toxicity. As we await results of randomized prospective trials, our meta-analysis supports the use of dose-intensive chemoimmunotherapy alone for the initial management of PMBCL. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Survival of Patients with Primary Mediastinal B-Cell Lymphoma Treated By Immunochemotherapy with or without Radiotherapy

Introduction: Primary mediastinal B-cell lymphoma (PMBL) is B-non-Hodgkin lymphoma subtype with unique clinical and biological characteristics mostly affecting younger patients. The current treatment standard is based on combination of rituximab and anthracycline based chemotherapy (R-CHT), either standard (R-CHOP) or intensified (DA-EPOCH-R). Radiotherapy (RT) consolidation has been standard of care for a long time, but its role is currently unclear, especially in the context of PET/CT guided therapy and long term toxicity prevention considering young patient population (Hayden at al.2020, Giulino-Roth 2018). In this analysis we evaluate the survival benefit of RT consolidation in patients with PMBL achieving PET negative (PET-neg) response after R-CHT. Methods: Patients with confirmed diagnosis of PMBL observed in NiHiL project (Gov Trial No: NCT03199066) were analysed. All patients were treated by R-CHT; PET/CT was performed at the end of the immunochemotherapy. The treatment response was evaluated according to Cheson 2007 criteria. Only patients achieving PET negativity were included in primary analysis. Results: The study included 230 patients out of whom 133 (57.8%) were women. Median age was 37 years (range 19-79); localised stage (I-II) was observed in 165 patients (71.7%) and low or low-intermed IPI in 189 patients (82.2%). The diagnosis was confirmed by central pathologist in all cases. Patients were treated by R-CHT, mostly R-CHOP (52.2%) and R-CHOEP-like (25.2%; i.e. R-CHOEP or DA-EPOCH-R); other intensive regimens, incl. platinum were used in 19.6% of patients; 48 patients (20.8%) received ASCT as part of first line induction. PET/CT was performed at the end of R-CHT (including ASCT); 174 patients (75.7%) achieved PET-neg remission at the end of the immunochemotherapy, 56 (24.3%) were PET positive (PET-pos). After reaching PET negativity, radiotherapy was performed in 68 patients (39.1%) with median dose of 36Gy (range 20-42Gy); 106 patients were not consolidated by RT and observed (OBS). There was higher incidence of bulky disease and B-symptoms in RT group (57% vs 47% and 57% vs 48% resp.) without reaching statistical significance. Median follow up was 5.9 years. OS and PFS probability at 6y were 90.5% and 83.6% resp. for the whole cohort. RT consolidation in PET-negative subgroup led to a significant improvement (p=0.039) in PFS at 6y with 95% (n=68) vs 85.3% in OBS group (n=106). This was, however, not transformed into the OS difference (94.5% vs 92.1% probability at 6y for RT vs OBS; HR 1.64, CI95% 0.52-5.8), figure 1. In subgroup of patients treated by R-CHOP, there was an insignificant trend for better PFS probability in RT group (HR 2.41, 95%CI 0.71-6.92, p=0.17). This was not reflected in any OS difference. The outcome of patients who reached PET negativity by R-CHOEP was the same regardless of the radiotherapy use. Conclusion: Our analysis of retrospectively recorded patients shows that skipping RT in PET-neg PMBL patients at the end of immunochemotherapy does not hamper overall survival. There was borderline improvement of PFS. The patient cohorts with different chemotherapy regimens were not big enough to analyse the RT significance. However, data observed in patient subpopulation treated by CHOEP-like regimen confirm already published observations (Dunleavy et al. 2013) concluding possible RT oblivion after intensified immunochemotherapy. Therefore, the choice between radiotherapy consolidation vs observation in PET negative patients should be discussed with the patient especially when less intensive regimen (R-CHOP) is used. Acknowledgement: This work was supported by Cooperatio Program - research area Oncology and Hematology and by research grant NU21-03-00411 Figure 1. Survival analysis of patients reaching PET negative remission after chemotherapy (n=174) shows improvement in progression free survival (PFS) for patients consolidated with radiotherapy (RT), however without reflection in overall survival (OS). Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

ABCL-371 STR Loci Allele Imbalance at 9p24.1 in Primary Mediastinal B-Cell Lymphoma

Molecular factors that determine the prognosis of immunotherapy response in primary mediastinal B-cell lymphoma (PMBCL) are not yet described. PD-L1/L2 expression levels in tumor samples have demonstrated clinical utility for many solid tumors. PD-L1/L2 expression levels could be affected by chromosomal lesions on 9p24.1 (losses or gains). Short tandem repeat (STR) analysis of appropriate markers is the simplest way to detect these lesions. To analyze possible associations between STR loci allele imbalance (AI) on 9p24.1 and PD-L1 expression levels in tumor cells from PMBCL patients. Two STR Loci (gt) n and (ttat)m on 9p24.1 in tumor cells were analyzed for a cohort of 36 PMBCL patients admitted to the National Medical Research Center for Hematology (Moscow, Russia) from 2013 to 2022. For six patients, STR data were compared with PD-L1 expression assessed by immunohistochemistry. DNA was isolated from tumor biopsy samples taken at diagnosis. Control DNA samples were taken from the blood of patients in complete remission or from the buccal epithelium. STR analysis for AI was performed by PCR with original primers. The fragment analysis was performed on an ABI3130 Genetic Analyzer. Inclusion criteria: de novo diagnosed PMBCL patients. pretreatment. The heterozygosity level for the (gt)n locus was 31 of 36 patients (86%), and AI was found in 16 of 31 (51.6%). The heterozygosity level for the (ttat)m locus was 20 of 36 patients (56%), and AI was found in 9 of 20 (45%). One patient was homozygous for both loci. No patients were found with AI at one locus and a normal heterozygous variant at the second locus. PD-L1 expression in tumor cells was evaluated for six patients. In two patients with aberrant STR loci, PD-L1 expression was found in 75%-100% of tumor cells. None of the four patients with normal 9p24.1 STR markers had PD-L1-positive cells. The high incidence of AI in the 9p24.1 locus indicates its possible involvement in the pathogenesis of PMBCL. The association between AI and PD-L1 expression in tumor cells requires confirmation in an extended cohort of patients. This study was supported by Rakfond grant 2/2020.

High efficacy of intensive immunochemotherapy for primary mediastinal B-cell lymphoma with prolonged follow up

Primary mediastinal B-cell lymphoma (PMBL) is currently curable in 85–95% of patients. Treatment regimens frequently used include RCHOP ± radiotherapy, DAEPOCH-R, or occasionally more intensive protocols. Here we present results of treatment of 124 patients with PMBL over a period between 2004 and 2017 with the use of a protocol designed for aggressive B-cell lymphoma GMALL/B-ALL/NHL2002 including 6 cycles of alternating immunochemotherapy with intermediate-dose methotrexate in each cycle, and reduced total doxorubicin dose (100 mg/m2 for whole treatment). Majority of patients (77%) received consolidative radiotherapy. A median (range) age of patients was 30 (18–59) years, and 60% were female. With a median (range) follow up of 9 (1–17) years, 5-year overall survival (OS) and 5-year progression free survival (PFS) were 94% and 92%, respectively. Positron emission tomography—computed tomography (PET-CT) results at the end of chemotherapy were predictive for outcome: OS and PFS at 5 year were 96% and 94% in PET-CT negative patients, respectively, and 70% and 70% in PET-CT-positive patients (p = 0.004 for OS, p = 0.01 for PFS). Eight (6%) patients had recurrent/refractory disease, however, no central nervous system (CNS) relapse was observed. Acute toxicity included pancytopenia grade 3/4, neutropenic fever, and treatment related mortality rate of 0.8%. Second malignancies and late cardiotoxicity occurred in 2.4% and 2.4% of patients, respectively. Intensive alternating immunochemotherapy protocol GMALL/B-ALL/NHL2002 is curative for more than 90% of PMBL patients and late toxicity in young patients is moderated. The attenuated dose of doxorubicin and intermediate dose of methotrexate may contribute to low incidence of late cardiotoxicity and effective CNS prophylaxis.

Circulating tumor DNA in primary mediastinal large B-cell lymphoma versus classical Hodgkin lymphoma: a retrospective study

Few data exist concerning circulating tumor DNA (ctDNA) relevance in primary mediastinal B-cell lymphoma (PMBL). To explore this topic, we applied a 9-gene next-generation sequencing pipeline to samples from forty-four PMBL patients (median age 36.5 years). The primary endpoint was a similarity between paired biopsy/plasma mutational profiles. We detected at least one variant in 32 plasma samples (80%). The similarity between the biopsy and ctDNA genetic profiles for the 30 patients with paired mutated biopsy/plasma samples was greater than or equal to 80% in 19 patients (63.3%). We then compared PMBL ctDNA features with those of a cohort of Hodgkin lymphoma patients (n = 60). The top three mutated genes were SOCS1, TNFAIP3, and B2M in both lymphoma types. PMBL displayed more alterations in TNFAIP3 (71.9% vs. 46.3%, p = 0.029) and GNA13 (46.9% vs. 17.1%, p = 0.013) than cHL. Our 9-gene set may delineate tumor genotypes using ctDNA samples from both lymphoma types.

Dose-adjusted EPOCH-R is not superior to sequential R-CHOP/R-ICE as a frontline treatment for newly diagnosed primary mediastinal B-cell lymphoma: Results of a bi-center retrospective study.

PurposePrimary mediastinal B‐cell lymphoma (PMBCL) is a rare subtype of diffuse large B‐cell lymphoma (DLBCL). Despite its aggressive course, PMBCL is considered curable. While in recent years dose‐adjusted (DA) EPOCH‐R (rituximab, etoposide, prednisone, vincristine, cyclophosphamide and doxorubicin) has become widely endorsed as first‐line therapy for newly‐diagnosed PMBCL, the optimal treatment for this disease and the role of radiotherapy (RT) remains unclear. DA‐EPOCH‐R provides good clinical outcomes, albeit is associated with short‐ and long‐term toxicity. To address this issue, the current retrospective bi‐icenter analysis compared efficacy and toxicity of DA‐EPOCH‐R and a less toxic R‐CHOP/R‐ICE regimen used for the treatment of newly‐diagnosed PMBCL.Patients and MethodsThe study included all patients with a histologically confirmed PMBCL diagnosis treated with DA‐EPOCH‐R or R‐CHOP/R‐ICE between 01/2013‐12/2020 at two tertiary medical centers. Patient demographic and clinical data were derived from institutional electronic medical records. The analysis included 56 patients: 31 received DA‐EPOCH‐R and 25 – R‐CHOP/R‐ICE.ResultsAt a median follow‐up of 1.9 years (IQR 3.1 years), similar progression‐free survival (2.1 versus 2.4 years; p = 0.7667), overall survival (2.5 versus 2.7 years; p = 0.8047) and complete response (80%) were observed in both groups. However, DA‐EPOCH‐R was associated with significantly longer hospitalization required for its administration (p < 0.001) and a trend for higher frequency of infections, stomatitis, thrombotic complications and febrile neutropenia‐related hospitalizations.ConclusionDA‐EPOCH‐R and R‐CHOP/R‐ICE provide similarly encouraging outcomes in newly‐diagnosed PMBCL patients. R‐CHOP/R‐ICE is associated with lower toxicity and significantly reduced hospitalization. Our findings suggest that this regimen may be considered as an alternative to DA‐EPOCH‐R in this patient population.

ABCL-261: STR Profiling Reveals Tumor Genome Instability in Primary Mediastinal B-Cell Lymphoma

<h3>Context</h3> Molecular factors to determine the prognosis for primary mediastinal B-cell lymphoma (PMBCL) are not described yet. In this regard, the loss of heterozygosity (LOH) in STR loci and elevated microsatellite alteration at selected tetranucleotides (EMAST) in the tumor genome are of interest. <h3>Objective</h3> To analyze the frequency of LOH and EMAST in tumor cells of PMBCL patients at diagnosis and to check possible association with disease features. <h3>Design</h3> STR profiles of the DNA of tumor cells were analyzed in a cohort of 72 PMBCL patients undergoing EPOCH regimen treatment at the National Research Center for Hematology (Moscow, Russia). <h3>Setting</h3> DNA was isolated from tumor biopsy samples taken at diagnosis. Control DNA samples were taken from the blood of patients in complete remission or from the buccal epithelium. STR profiles for LOH and EMAST analysis were assessed by PCR with the COrDIS Plus multiplex kit for amplification of 19 STR markers and amelogenin loci (Gordiz Ltd, Russia). The fragment analysis was performed on an ABI3130 Genetic Analyzer. <h3>Patients</h3> Inclusion criterion: <i>de novo</i>-diagnosed PMBCL patients. Exclusion criterion: pre-treatment. <h3>Main Outcomes Measures</h3> Progression-free survival (PFS) and overall survival (OS) at 24 months from diagnosis were used to assess the independent impact of genetic instability as a risk factor. <h3>Results</h3> LOH was detected in 37 of 72 patients (51.4%). EMAST was found in 40 patients (55.5%); 24 had a combination of EMAST with LOH. For all EMAST-positive patients, the number of involved loci was no more than 4 out of 19 (EMAST-low, less than one-third of the panel of microsatellite markers). It was shown that OS and PFS were the highest in patients with a stable STR tumor profile. <h3>Conclusions</h3> STR profiling of PMBCL tumor genome revealed LOHs associated with deletions or uniparental disomy and detected EMAST attributed to failure of DNA mismatch repair. The high incidence of LOH and EMAST in PMBCL can be associated with standard therapy failure in a significant number of patients. <h3>Funding</h3> This study was supported by Rakfond grant 2/2020.

Outcomes of Patients with Primary Mediastinal B-Cell Lymphoma Treated with Dose Adjusted R-EPOCH Regimen: A Single Centre Experience.

Primary Mediastinal B cell lymphoma (PMBCL) is a biologically and clinically distinct subset of diffuse large B cell lymphoma. We analysed the outcomes of our cohort of PMBCL patients treated with Dose adjusted (DA)-R-EPOCH regimen. This is a retrospective analysis of consecutive PMBCL patients who received chemotherapy consisting of DA-R-EPOCH with filgrastim support. Survival analysis was done using Kaplan-Meier method. All calculations were performed using SPSS version 20 for windows. A total of 43 consecutive suspected PMBCL patients were reviewed for this study, 6 patients were excluded as diagnosis of PMBCL could not be established. All patients except one (97.3%) received 6 cycles of R-DA-EPOCH regimen. Median age of the patients was 27years (range 15-58). Bulky disease (> 7cm) was present in 97% patients and 54% patients had extranodal disease. With a median follow up of 40months, 3-year overall survival was 80.6% (95% CI: 74.0-87.2). The 3-year event free survival was 78.4% (95% CI: 71.6-85.2). There were 6 (16.2%) relapses, 1 (2.7%) primary progression and 7 (23%) deaths. Mediastinal radiotherapy was administered to 17 (45.9%) patients. All the deaths were due to disease progression. Grade III/IV toxicities were seen in 28 (75.7%) patients, febrile neutropenia being the most common one. DA-R-EPOCH regimen is an effective and tolerable regimen in PMBCL patients even with adverse features.

Ctdna Monitoring in Predicting Relapse of Primary Mediastinal B-Cell Lymphoma

Background: Primary mediastinal B-cell lymphoma (PMBCL) is a rare subtype of aggressive B-cell lymphoma. The majority of relapses occur within the first few months resulting in a dismal prognosis, thus we need to identify the risk of early relapse. Circulating tumor DNA (ctDNA)-based response assessment may enable response adapted therapy and early prognostication. In this study, we aim to clarify the role of ctDNA-monitoring in early prediction of relapse. Methods: We retrospectively analyzed the characteristics of 38 patients with PMBCL. The genetic sequencing and ctDNA monitoring were conducted by target next-generation sequencing (NGS). Results: The median age was 32 years old and 26 (68.4%) were female. Most cases (31/38, 81.6%) were diagnosed with stage I or II disease and fourteen (36.8%) cases present with extranodal involvement. The 5-year PFS and OS for PMBCL patients were 65.7 and 72.1%, respectively. High IPI score were associated with poor survival by univariate analysis. The complete response (CR) rate and overall response rate of R-CHOP was similar to R-EPOCH (69.2% vs. 70.0%, P =1.000 and 84.6% vs. 95.0%, P =0.547, respectively), however the 5-year PFS and OS rate for R-EPOCH seems longer than those for the R-CHOP (PFS 82.9% vs. 53.8%, P=0.0569; OS 92.9% vs. 60.6%, P=0.0567). Radiotherapy was less delivered in the patients who received R-EPOCH than R-CHOP (5.0% vs.53.8%, P=0.003) and had no impact on prognosis. We performed genetic sequencing on twenty-three cases. STAT6(15/23, 65.2%), SOCS1(13/23, 56.5%), TNFAIP3(13/23, 56.5%) were the most common affected genes in both response and refractory/relapsed patients. Cell cycle, FoxO signaling pathway and TNF signaling pathway were more frequently involved in refractory patients. ctDNA monitoring was conducted in 16 cases and fifteen cases show undetectable ctDNA after a median of 2 (range from 1 to 5) cycles and one refractory case presented with durable positive ctDNA. Among 4 relapse cases, 3 had prior detectable ctDNA half a month earlier before imaging manifestations of relapse. Regarding prediction of relapse, positive predictive value of ctDNA is 100%. Conclusions: Intensive induction chemotherapy can improve prognosis of PMBCL and ctDNA precisely predicted relapse. As most cases relapse within a few months, ctDNA-monitoring is crucial to improve prognosis. Disclosures No relevant conflicts of interest to declare.

Checkpoint Inhibition before Axicabtagene Ciloleucel Cell Therapy in Primary Mediastinal B-Cell Lymphoma (PMBCL) Treated in Real Life Setting

Background: Eighty-five percent of PMBCL are cured by standard therapy, but the outcome of refractory/relapsed (R/R) PMBCL is very poor. Checkpoint inhibitors (CPIs) have shown promising activity in relapsed PMBCL. Axibactagene ciloleucel (axi-cel) CAR-T cell therapy, can induce durable responses and is currently approved for the treatment of adult patients with R/R PMBCL. Aims of this analysis were: to register all Italian PMBCL patients candidate to CAR-T in the 6 active centers;to evaluate the intention to treat overall response rate (ORR, complete [CR] and partial response [PR]) in patients treated with axi-cel and CPIs for salvage or bridging before CAR-T and for relapse after CAR-T;to evaluate cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS). Methods: In August 2019 the Italian Drug Agency (AIFA) approved axi-cel; before the reimbursement by AIFA, an expanded access program supported by Kite/Gilead started. One patient slot per month per qualified center was available. Patients were included in a large national CAR-T prospective observational study approved by ethics committees. Results: Since April 2019 to March 2020, 20 R/R PMBCL were evaluated and 18 were apheresized in order to receive axi-cel; 2 were excluded because active CNS disease in one, and eligibility to transplant, while in CR, in the second one. Their clinical characteristics were: median age 38 years (range 22-50), male 8 (44%), stage II 6 (33%), advanced stage III/IV 12 (66%), bulky disease 6 (33%); LDH upper than normal 3 (2%). Median number of prior lines was 3 (2-6); 5 patients (28%) had a previous autologous stem cell transplant and 12 (66%) received a prior radiotherapy. The majority of patients, 16 (89%) were refractory to the last treatment when they were evaluated for CAR-T eligibility; 9 of 18 patients had CPI exposure before leukoapheresis: 6 pembrolizumab and 3 nivolumab in combination with brentuximab-vedotin. No manufacturing failures were reported. Bridging therapy was performed in 16 of 18 patients (88%). Seventeen patients (94%) received lymphodepleting Flu-Cy chemotherapy and only 16 pts received CAR-T for central nervous system (CNS) progression during bridging therapy (n=1) and respiratory failure due to pneumonia (n=1); the 2 patient not infused were exposed to CPIs. Median vein to vein time was 40 days (30-79). Median follow-up time for infused patients was 209 days (9-444). CRS was observed in 12 of 16 infused patients: 5 grade 2 and 7 grade 1. ICANS (2 grade 1, 2 grade 2, 1 grade 3) was recorded in 5 patients. No differences regarding CRS and ICANS occurrence were observed in patients exposed or not to CPIs. At 30-days after the infusion, all the 16 infused patients were evaluable for response: 7 (44%) CR, 5 (31%) PR, with ORR 75%, 3 (19%) stable disease (SD) and 1 (6%) progressive disease (PD). Two patients in PR at 30 days converted to CR at 90 days, with continuous CR at 180 days; all the 3 patients in SD and 1 out of 5 in PR at 30 days progressed at 90 days. Considering the 9 patients exposed to CPIs before CAR-T, 7 out of 9 were infused and all the 7 infused were evaluable for response: 2 (29%) CR, 4 (57%) PR, with ORR 86%, and 1 (14%) died because of a rapid CNS progression after infusion. Two patients in PR at 30-days converted to CR at 90-days, one with continuous CR at 180 days after CAR-T. Conclusions: In our series of 16 infused patients, axi-cel was effective with an ORR of 75% (CR 44%) at 30-days after CAR-T infusion and ORR of 54% (CR 46%) in the 13 patients evaluable at the median follow-up time (180-days after CAR-T infusion). It is important to note the 4 patients from the original real life cohort never received axi-cel. It is noteworthy that ORR was 86% in patients receiving CPIs before CAR-T and 75% in those not exposed to CPIs. With the limitation of small number, the exposure of immune-checkpoint inhibitors seems not to affect negatively response rate and adverse events. Disclosures Chiappella: Janssen: Honoraria; Iqone: Honoraria; Servier: Honoraria; Roche: Honoraria; Celgene: Honoraria; Gilead-Kite: Honoraria; Takeda: Honoraria. Zinzani:Bayer: Consultancy. Corradini:BMS: Other; Sanofi: Consultancy, Honoraria; Servier: Consultancy, Honoraria; Takeda: Consultancy, Honoraria, Other; Amgen: Consultancy, Honoraria, Other: Travel and accommodations paid by for; AbbVie: Consultancy, Honoraria, Other: Travel and accommodations paid by for; KiowaKirin: Consultancy, Honoraria; Novartis: Consultancy, Honoraria, Other: Travel and accommodations paid by for; F. Hoffman-La Roche Ltd: Consultancy, Honoraria; Janssen: Consultancy, Honoraria; Incyte: Consultancy; Gilead: Consultancy, Honoraria, Other: Travel and accommodations paid by for; Daiichi Sankyo: Consultancy, Honoraria; Celgene: Consultancy, Honoraria, Other: Travel and accommodations paid by for; Kite: Consultancy, Honoraria.

ABCL-349: Genetic Instability of Microsatellite Loci in Primary Mediastinal B-cell Lymphoma

Context: Loss of heterozygosity (LOH) in STR loci, elevated microsatellite alteration at selected tetranucleotides (EMAST) and instability of mononucleotide tandem repeats (MSI - microsatellite instability) are intrinsic features of diverse tumor cells. For primary mediastinal B-cell lymphoma (PMBCL) these molecular factors to determine the prognosis of the disease have not been described yet. Objective: To analyze the frequency of LOH, EMAST, and MSI in tumor cells of PMBCL patients at diagnosis and to check possible association with disease features. Design: STR profiles of the DNA of tumor cells were analyzed on a cohort of 36 PMBCL patients undergoing treatment at the National Research Center for Hematology (Moscow, Russia). Setting: DNA was isolated from tumor biopsy samples taken at diagnosis. Control DNA samples were taken from the blood of patients in complete remission or from the buccal epithelium. STR-profiles for LOH and EMAST analysis were assessed by PCR with COrDIS Plus multiplex kit for amplification of 19 STR-markers and amelogenin loci (Gordiz Ltd, Russia). COrDIS MSI (BAT-25, BAT-26, NR-21, NR-24 and NR-27) kit was used for MSI testing. The fragment analysis was performed on ABI3130 Genetic Analyzer. Patients: Inclusion criteria: de novo diagnosed PMBCL patients. There were 10 males and 26 females, the median age at presentation was 30 years (range 21–61). Main outcome measures: Progression-free survival (PFS) and overall survival (OS) at 24 months from diagnosis were used to assess the independent impact of the genetic instability as a risk factor. Results: LOH was revealed in 21 out of 36 patients (58%). EMAST was found in 18 patients (50%); in 12 of them EMAST was accompanied by LOH, and in one by MSI. MSI was found in three patients only. OS and PFS were highest in the group with stable STR profile of a tumor, but not formally significant because of the small sample size. Conclusions: The high incidence of LOH and EMAST in PMBCL can cause standard therapy failure in a significant number of patients. EMAST as a variant of microsatellite instability might be useful for further consideration as a predictor of response to checkpoint inhibitor therapy.

Highly Favorable Outcomes with Salvage Radiation Therapy Followed By Autologous Transplant in Relapsed and Refractory DLBCL Patients with Minimal or No Response to Salvage Chemotherapy

Introduction: For patients with relapsed or primary refractory (rel/ref) diffuse large B-cell lymphoma (DLBCL) who respond to salvage chemotherapy, high-dose chemotherapy and autologous hematopoietic cell transplantation (HDT-AHCT) is considered standard of care. Patients with refractory disease to salvage chemotherapy, defined as stable disease (SD) or progressive disease (PD), by functional imaging are ineligible for HDT-AHCT, and have a poor prognosis. In practice, we have attempted to salvage these patients with radiation therapy (RT) to residual sites of active disease prior to consolidative HDT-AHCT. The outcome of this unique combined modality salvage paradigm has not been previously reported. Methods: We retrospectively reviewed all patients with rel/ref DLBCL who received salvage chemotherapy followed by salvage RT and HDT-AHCT between the years of 2000 and 2017 at a single center. Only patients with SD or PD as defined on the 5-point Deauville scale after salvage chemotherapy and who had at least 1 year of follow-up were included in this analysis. The second-line age-adjusted International Prognostic Index (sAAIPI) was determined at the time of initiation of salvage chemotherapy.Survival functions were estimated by the Kaplan-Meier method and compared using a log-rank test. Results: Thirty-six patients, 12 with relapsed and 24 with primary refractory disease, with a median age of 44 years (range: 19-68 years) were analyzed. Twenty-three patients had DLBCL while 13 had primary mediastinal B-cell lymphoma (PMBCL). The majority of patients had KPS 80-100 (n=32, 89%), 0-1 extranodal sites (n=30, 83%), and normal LDH (n=21, 58%). The sAAIPI scores for this cohort were as follows: 0 (n=10), 1 (n=21), 2 (n=4), and 3 (n=1). All patients received salvage chemotherapy with subsequent functional imaging showing SD (n=32) and PD (n=4) and then went on to receive salvage RT to the sites of active disease. Median RT dose was 39.6Gy (range: 30-54Gy). Six patients also received TBI as part of their conditioning regimen prior to HDT-AHCT. With median follow up of 4.0 years (range: 1.0-12.3 years) for survivors, 4-year relapse-free survival (RFS) was 75.6% and 4-year overall survival (OS) was 80.3% (Figure 1a). There was no significant difference in 4-year RFS for patients with relapsed versus primary refractory disease (80.2% vs 74.8%, p=0.59). PMBCL patients had better RFS than DLBCL patients (92.3% vs 67.8%, p=0.12). Using the composite sAAIPI score was highly prognostic with worse outcomes for patients with higher risk sAAIPI scores. By sAAIPI score, 4-year RFS was 80.0% for a score of 0, 90.2% for a score of 1, and 0% for scores of 2 and 3. Patients with low- and low-intermediate risk sAAIPI scores of 0 and 1 had improved RFS as compared to patients with sAAIPI scores of 2 and 3 (87.0% vs 0%, p&lt;0.0001 (Figure 1b). Conclusions: Patients with chemorefractory rel/ref DLBCL who have had minimal or no response to systemic salvage therapy may benefit from salvage RT to the residual PET-avid disease followed by HDT-AHCT, particularly if their sAAIPI score is ≤ 1. The outcome of this retrospective cohort is markedly superior to outcomes described in the literature for this high-risk population and represents a promising treatment paradigm to be further explored. Emerging data suggest similar patients may benefit from CAR T-cell therapy. Given the limited availability and high cost of CAR T-cell therapy, we suggest there may be a role for sequencing this combined-modality salvage paradigm prior to CAR T-cell therapy in order to provide these poor-risk patients with an additional line of therapy. Disclosures Scordo: Angiocrine Bioscience, Inc.: Consultancy; McKinsey &amp; Company: Consultancy. Sauter:Sanofi-Genzyme: Consultancy, Research Funding; GSK: Consultancy; Spectrum Pharmaceuticals: Consultancy; Novartis: Consultancy; Genmab: Consultancy; Precision Biosciences: Consultancy; Kite/Gilead: Consultancy; Celgene: Consultancy; Juno Therapeutics: Consultancy, Research Funding.