Citations

Introduction: In recent years, increasing evidence suggests an important role of tumor-infiltrating B lymphocytes (TIL-Bs) in cancer immunity. In various types of solid tumors, high density of TIL-Bs and presence of mature tertiary lymphoid structures consistently correlate with better prognosis and response to immune checkpoint blockade therapies. Diffuse large B-cell lymphoma (DLBCL) is a type of aggressive lymphoma arising from mature B cells. TIL-Bs have been identified in DLBCL based on single-cell gene-expression data in several previous studies, but their prognostic significance in DLBCL is unknown. In this study, we determined the clinical significance of TIL-B abundance and investigated the potential underlying mechanisms. Patients and Methods: This abstract includes a discovery study in which we distinguished TIL-Bs from malignant B-cells by immunoglobulin gene (IG) clonotype analysis and a validation study using flow cytometry. In the discovery study, a molecularly well-characterized DLBCL cohort assembled by the International DLBCL R-CHOP Consortium Program was analyzed. DNA samples from DLBCL patients were sequenced by high-throughput ultra-deep sequencing assays using multiplexed forward and reverse primers. Based on the IG heavy chain (IGH) sequencing results, malignant B-cell and TIL-B VDJ clonotypes were identified for 138 patients, and the frequencies and numbers of TIL-B clonotypes in individual sequence repertoires were calculated. In 79 DLBCL cases, a DJ-only sequence was predominant and no productive VDJ sequences met the criteria for diagnostic sequences. Total frequencies of productive VDJ sequences in the sequence repertoires were then used to access high and low TIL-B frequencies. TIL-B frequencies were also evaluated by IG light chain clonotypes for these 217 patients and additional 52 patients. The TIL-B frequencies were correlated with patient survival, gene-expression profiling (GEP) data GSE31312, and frequencies of DLBCL-infiltrating immune cells quantified by fluorescent multiplex immunohistochemistry (mIHC). To validate the prognostic significance of TIL-B frequencies, we assembled a validation cohort of 94 DLBCL patients treated at Duke University Medical Center. TIL-Bs in tissues were quantified by flow cytometry immunophenotyping. Ratios of TIL-Bs to abnormal B cells were assessed for prognostic effects. Results: First, we found thathigh frequencies of TIL-Bs and clonotypic diversity were associated with significantly favorable prognostic effects in 138 DLBCL patients with their malignant B-cell IGH-VDJ clonotypes identified (Figure A). The prognostic effects of TIL-Bs outweighed those of the germinal center B-cell-like/activated B-cell-like classification (Figure B). Second, patients with high frequencies of TIL-Bs compared with those without had significant upregulation of immune response genes by GEP (Figure C), higher CD4 T cell infiltration quantified by fluorescent mIHC (Figure D), and enrichment of EcoTyper-defined specific cell states, including significantly increased frequencies of B cell state S2 (pre-memory), CD4 T cell state S3 (naïve), follicular helper T cell state S3, and endothelial cell state S3, whereas lower frequencies of B cell state S5 (plasmablast) and follicular helper T cell state S1. Third, high TIL-B abundance in 79 ‘DJ-only’ clonotype cases and additional 52 cases assessed by light chain clonotype analysis showed similar favorable prognostic effects and associations with upregulated immune gene expression and CD4 T cell infiltration. Fourth, the identified GEP signatures, prominent with genes involved in T-dependent B cell activation and T cell activation and differentiation, showed significant prognostic effects in two GEO datasets. Fifth, TIL-B abundance showed significant prognostic effects in DLBCL cases infiltrated with T-cells in both the discovery study (Figure E) and validation study (Figure F, flow cytometry cohort). Conclusions: TIL-B frequencies and clonotypic diversity have remarkable prognostic and biological effects in DLBCL, which has significant implications for developing novel therapeutic strategies. Our study identified a new prognostic biomarker in DLBCL and an important player in anti-lymphoma immune responses, providing a new point of view to analyze the tumor biology in aggressive B-cell lymphoma and precisely identify prognostic determinants.

Introduction: AutoHCT is an effective therapy for patients (pts) with DLBCL who are refractory or relapse following first-line chemotherapy, but long term outcomes remain suboptimal. The BTK inhibitor ibrutinib has shown promising activity in pts with ABC-DLBCL (Wilson et al, Nat Med 2015). In 2016 we launched a randomized phase III clinical trial to evaluate the addition of ibrutinib to AutoHCT as part of conditioning and as consolidation therapy for pts with relapsed/refractory ABC-DLBCL. Patients and Methods: Following assessment of open-label treatment in a safety cohort of 6 pts, we randomly assigned pts with chemotherapy-sensitive persistent or relapsed ABC-DLBCL to receive ibrutinib 560 mg [Arm A] or placebo [Arm B] concurrently with conditioning on days -6 to -1 of AutoHCT (Cycle 1). Following count recovery and resolution of acute toxicities, pts continued ibrutinib 560 mg in Arm A or placebo in Arm B for 12 additional 28-day cycles. Crossover to Arm A using single agent ibrutinib was allowed for pts progressing on Arm B. Central pathology review was performed for disease confirmation and subtype assignment. Initially, ABC subtype assignment was performed using the Nanostring Lymphoma Subtyping Test (LST) but the study was amended to use the IHC-based Hans classifier for non-GCB on 8/15/19. Other key eligibility criteria included receipt of no more than 3 prior regimens, no active CNS involvement, no need for long-term anticoagulation, and no prior progression on Ibrutinib. Pts were stratified by time to relapse (<=12 mos vs >12 mos), type of conditioning (BEAM or CBV) and prior ibrutinib use. The original design required 296 patients with 82% power to detect a difference in 2-year progression-free survival (PFS) rates of 67% vs. 50% with a two-sided alpha of 0.05. A revised statistical design allowed an interim futility analysis to be conducted after 40 pts were enrolled and followed for 2 years. Even though the study met criteria to continue accrual to a target of 160 pts, we decided to close it early due to declining accrual on 12/20/2022. We hereby present the primary analysis of the evaluable patients enrolled (data as of 07/17/2023). Results: Between 2/6/17 and 8/24/18, we screened 255 pts but enrolled only 34 (13%) largely due to long assay turnaround times and performance characteristics with the LST. Following the switch to the IHC-based Hans classifier (after 8/15/19), 60 out of 110 pts screened were enrolled (54%) in line with our a priori expectation. In total, 45 pts were randomized to Arm A vs. 43 to Arm B, with 39 and 38, respectively, being evaluable for the primary endpoint at this time. Median age was 61 years (range 20-75 years), median number of prior lines of therapy was 2, 46% of the pts were female, 21% were refractory or had early relapse, 94% were planned to receive BEAM conditioning. Overall, 87% of pts in Arm A and 88% in Arm B underwent conditioning while 70% and 67% respectively, started maintenance therapy. The median number of cycles received was 6 in Arm A vs. 5 in Arm B. The primary endpoint of 2-year PFS was 57.6% (95% CI: 42.1%, 78.8%) in Arm A vs. 40.8% (95% CI: 26.5, 62.8%) in Arm B (HR=0.54; 95% CI: 0.26, 1.10; log-rank p = 0.087). Median PFS was 26.5 months (95% CI: 10.2, not reached [NR]) in Arm A vs. 8.1 months (95% CI: 5.4, NR) in Arm B. Only one pt crossed over to Arm A after progression on Arm B. The most common grade 3-4 AEs in each of the arms were: neutropenia (A: 14%, B: 14%), febrile neutropenia (A: 11%, B: 9%), and thrombocytopenia (A: 11%, B: 7%). Rates of any grade atrial fibrillation were similar between arms (12.8% in Arm A vs. 10.5% in Arm B); one patient had grade 3 and one patient had grade 4 atrial fibrillation, both in Arm B. Six pts had reported grade 5 events (Arm A: 2 sepsis, 1 COVID-19, 1 respiratory failure and 1 disease progression and Arm B: 1 disease progression) but only one grade 5 event (sepsis, Arm A) was deemed probably related to treatment. Conclusions: Despite an early high rate of screen failures and declining accruals due to competition with CAR-T cell therapy, the primary analysis suggests that ibrutinib with and following AutoHCT in relapsed/refractory ABC-DLBCL is well-tolerated and may improve PFS in this high-risk patient population. This trial was supported by: U10CA180821, U10CA180882, U24CA196171; U10CA180820, UG1CA233234 (ECOG-ACRIN); U10CA180868 (NRG); U10CA180888 (SWOG) ClinicalTrials.gov Identifier: NCT02443077 https://acknowledgments.alliancefound.org

Introduction: Chronic lymphocytic leukemia and small lymphocytic lymphoma (CLL/SLL) are the most prevalent types of leukemia in the United States. Given the emergence of novel CLL/SLL therapies, it is important to understand real-world treatment characteristics and unmet needs. The objectives of this retrospective cohort study were to examine patient characteristics, treatment patterns, and associated outcomes for patients with CLL/SLL, overall, and by CLL and SLL, respectively. Methods: A retrospective study was conducted using Symphony Integrated Dataverse (IDV ®), a comprehensive, longitudinal, open-claims database, and integrated electronic medical record data. Patients aged ≥18 years with ≥1 diagnosis for CLL or SLL and who initiated a first-line or second-line treatment between January 1, 2020, and December 31, 2022 (index period) were included in the study. Index date was defined as the date of treatment initiation. Patients were required to be continuously enrolled for 365 days prior to the index date and ≥90 days following the index date. Patient data were included until patients were lost to follow-up or to the end of the study period (March 31, 2023). Treatment patterns were examined with the following treatment regimens and stratified by line of therapy: chemotherapy (including bendamustine-based and other chemotherapy), anti-CD20-based, Bruton tyrosine kinase inhibitors (BTKiibrutinib- and acalabrutinib-based), and venetoclax-based therapy. Zanubrutinib use was not captured in the BTKi treatment category as its CLL/SLL indication approval date (January 19, 2023) was outside the index period. Discontinuation rate was defined as the percentage of patients who discontinued their current treatment regimen within 90 days. Healthcare resource utilization was measured by number of outpatient visits, inpatient visits, and other medical/hospital services per patient per month (PPPM) during the treatment regimen. Results:There were 17,748 and 5,667 patients with CLL/SLL who initiated a first-line or second-line treatment regimen during the index period, respectively. The most frequently used treatment regimens in first line were BTKi (49.9%), followed by anti-CD20 (26.1%), venetoclax (15.4%), and chemotherapy (8.6%). BTKi continued to be the predominant treatment (49.6%) in second line, followed by venetoclax (30.9%), chemotherapy (10.4%), and anti-CD20 (9.2%). First-line patients treated with BTKi regimens were older and Medicare-insured when compared with patients receiving venetoclax-based regimens (65+ years: 77.1% vs 66.9%, P-value <0.001; Medicare: 61.7% vs 53.0%, P-value <0.001). Patients receiving BTKi had significantly lower discontinuation rates compared with other treatment regimens during both first and second lines of therapies (Figure 1). Mean PPPM outpatient visits, inpatient visits, and other medical/hospital services were less frequent with BTKi therapies vs other treatment regimens across both lines of therapy (Table 1). Of the 17,748 patients initiating first line of therapy, 13.3% had SLL while the remaining 86.7% had CLL. The proportion of patients with SLL vs CLL decreased in second line and second line or later (8.0% vs 92.0% and 6.8% vs 93.2%, respectively). Treatment discontinuation rate in SLL patients was higher than that in CLL patients in first line (32.41% and 25.08%, respectively); however, discontinuation rates became comparable with later lines of therapy. Across all lines of therapy, healthcare resource utilization was significantly higher in SLL patients than CLL patients. Median PPPM outpatient visits were 1.02 for SLL and 0.67 for CLL. Median PPPM for other medical/hospital services was 0.7 for SLL and 0.23 for CLL. Conclusions: BTKi therapy, the primary treatment regimen across first- and second-line therapies, has significantly lower discontinuation rate and healthcare resource utilization compared with other treatment regimens. Further studies are needed to evaluate real-world clinical outcomes of CLL/SLL regimens to support evidence-based treatment decisions.

Background Low ferritin levels have been previously described to lower HbA 2 percentages on hemoglobinopathy evaluation. At our institution we have incorporated serum ferritin levels into an expanded panel test for thalassemia and hemoglobinopathy to assist with interpretation. There have been many critiques on the reference ranges used to define low ferritin in various pediatric ages. Lower ferritin levels, even in the absence of anemia or microcytosis, have been ascribed to nonspecific symptoms such as fatigue, weakness, difficulty concentrating, and decreased productivity that could be misconstrued as an attention-deficit disorder, especially in children. A hemoglobinopathy evaluation is often the first occasion for ferritin testing in patients; therefore, the hemoglobinopathy evaluation can serve as a key checkpoint to detect iron deficiency allowing for prompt treatment and amelioration of symptoms. Objective The purpose of this study was to evaluate the relationship between ferritin levels and the occurrence of hemoglobinopathies among pediatric patients in a single institution with the expectation that patients obtaining a hemoglobinopathy workup would be more likely to have a low ferritin (null hypothesis: there is no difference in the ferritin levels between patients receiving a hemoglobinopathy workup). The second hypothesis was that hemoglobinopathy testing would be done in a stepwise manner so at least 50% of patients obtaining a hemoglobinopathy workup would have prior interventions such as iron supplementation (null hypothesis: patients obtaining hemoglobinopathy evaluation have less prior interventions such as iron supplementation). Design/Method This retrospective, single center study included patients aged 0-18 years who had their first thalassemia and hemoglobinopathy testing for any indication between January 1, 2023 and July 28, 2023. Clinical variables were collected via review of the medical record. Low ferritin was defined as a ferritin level less than or equal to 25 mcg/L. Pearson's Chi Squared test and the Kruskal Wallis test were used for statistical analysis of categorical and continuous variables respectively. Results Seventy-seven total eligible patients were identified. Of those patients 35 were found to have low ferritin levels (45.5%). Among the patients with a low ferritin level, 15 patients also had a hemoglobinopathy present on their testing (41.7%) but there was no significant difference compared to those with normal ferritin levels (p value = 0.467). The patients with normal ferritin levels were of a significantly higher age compared to those with low ferritin levels 12.2 years vs 6.8 years (p value = 0.001) and 14 (42.4%) patients with low ferritin had no anemia and 29 (82.9%) never received prior iron supplementation. It was also noted that patients of Asian race were significantly more likely to have received prior iron supplementation 27.3% (p value = 0.004). Among patients with a hemoglobinopathy present the average age of their diagnosis was 9.1 years (95% Confidence Interval 6.6-11.7 years). Nine of the 10 patients with no hemoglobin values available were tested due to college sports participation and only 2 of those patients were found to have low ferritins. Conclusion Among our group of pediatric patients who underwent evaluation for thalassemia and hemoglobinopathy, there was no significant difference between those having a hemoglobinopathy present and lower or higher ferritin levels. Iron deficiency was more often noted among the younger patients even in the absence of anemia. It was important to note the majority of patients with iron deficiency detected on the hemoglobinopathy evaluation had never received prior iron supplementation until the time of their hemoglobinopathy evaluation. This also demonstrates that many providers may not do a stepwise workup but order all the evaluations at once. As there are many nonspecific symptoms of iron deficiency that can hinder the development of a child it is valuable to detect iron deficiency on hemoglobinopathy testing in order to intervene as soon as possible as iron deficiency can be masked by normal hemoglobin and normocytosis. Further investigation with larger study groups is warranted to better elucidate this relationship. Overall, our results suggest that enhanced characterization of iron deficiency parameters for patients receiving a hemoglobinopathy evaluation is warranted.

Background: Richter Transformation (RT) refers to the development of an aggressive lymphoma in the setting of chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL). Small molecule inhibitors (SMI), such as the Bruton tyrosine kinase inhibitors (BTKi) and BCL2 inhibitors (BCL2i), have revolutionized the treatment of CLL/SLL. Prior studies evaluated outcomes of patients (pts) with RT in an era where chemoimmunotherapy (CIT) was typically used to treat CLL/SLL. Now that SMI are standard of care, fewer pts with CLL/SLL receive CIT, therefore we sought to determine which variables predict survival in pts who developed RT without prior CIT exposure for CLL/SLL. Methods: We conducted an international multicenter retrospective study of pts from 11 academic centers. Pts with RT occurring as large B-cell lymphoma who did not previously receive CIT (e.g., fludarabine, cyclophosphamide, bendamustine, chlorambucil) for their CLL/SLL were included. We collected pt, disease, and treatment (tx) characteristics. RT was categorized into 3 groups: “Concurrent RT,” defined as RT and CLL/SLL diagnosed simultaneously (within 3 months [mos]); “RT w/o prior CLL/SLL tx,” defined as RT and CLL/SLL diagnosed >3 mos apart with the CLL/SLL never treated; and “RT with prior non-CIT tx for CLL/SLL” defined as having received prior CLL/SLL tx. Overall survival (OS) was measured from RT diagnosis (Dx) and estimated using the Kaplan-Meier method. Cox regression model was used to identify prognostic factors associated with OS. Results: 242 pts were identified. At CLL/SLL Dx, 66% of pts had unmutated IGHV, 37% had del17p/ TP53 disruption, and 20% had trisomy 12. At RT Dx, 56% of the RT with prior non-CIT tx for CLL/SLL group had a del17p/ TP53 disruption versus 27% for the concurrent RT group and 18% for the RT w/o prior CLL/SLL tx groups. Median time from CLL/SLL Dx to RT Dx was 47 mos (range 3-394) for RT w/o prior CLL/SLL tx and 46 mos (range 4-218) for RT with prior non-CIT tx for CLL/SLL. Pts with prior non-CIT tx for CLL/SLL received a median of 1 (range 1-5) prior tx. 86% of pts previously received a BTKi or a BCL2i for tx of CLL/SLL (54% BTKi only, 5% BCL2i only, 27% both). The remaining 14% had been treated with an anti-CD20 monoclonal antibody (MoAb), steroids, lenalidomide, radiation, or alemtuzumab. 85% of pts developed RT while on active tx with SMI. Median age at RT Dx was 69 years (range 37-92). At RT Dx, 38% and 21% of pts had del17p/ TP53 disruption, and trisomy 12 in CLL/SLL tissue, respectively. At RT Dx, there was no difference in highest SUV on PET (median 16, range 2.9-65.6), LDH (median 325, range 103-8162, ULN range 180-271), largest lymph node in diameter (median 4.9 cm, range 0-17.1), Ki-67 (median 80%, range 5%-100%), rate of GCB subtype (27%), presence of MYC translocation (20%), or CLL/SLL and RT clonal relationship (81%) for those pts tested amongst the 3 groups ( Table). The most common 1 st line tx for RT was CIT with or w/o radiation and/or steroids (76%), followed by CIT plus SMI (13%), SMI single agent or combination (7%), and other (4%, which includes anti-CD20 MoAb). Median follow-up from RT Dx was 42.3 mos; the median OS for the entire cohort was 25.8 mos (95% CI: 16.8-49.1). Pts with RT after non-CIT tx for CLL/SLL had a significantly worse OS (median 8.2 mos [95% CI: 5.6-14.3]) than pts with concurrent RT (median 46.3 mos [95% CI: 23.8-96.9]) and RT w/o prior CLL/SLL tx (median 63.5 mos [95% CI: 28.3-70.3]) ( Figure). From an exploratory univariable analysis, we found RT after non-CIT tx for CLL/SLL compared to concurrent RT (HR 2.32 [95% CI: 1.55-3.46]), RT Dx age (HR for 5-year increase 1.14 [95% CI: 1.05-1.23]), del17p/ TP53 disruption (HR 2.27 [95% CI: 1.45-3.56]), and LDH (HR for 2-fold increase 1.37 [95% CI: 1.20-1.58]) were prognostic for worse OS. Conclusions: This is the largest cohort of pts who developed RT having received only SMI with no CIT for their CLL reported, thus reflecting the current treatment landscape. Despite being CIT-naïve, pts in our series who received tx for their CLL/SLL and then developed RT had a short OS, underscoring the need to develop better therapies for these pts. In contrast, as in other published analyses (Wang Haematologica 2020), pts w/o prior tx for CLL/SLL had more favorable outcomes, with a median OS of approximately 5 years. Further analysis to determine PFS and OS by tx for RT for pts who have been treated with SMI for their CLL/SLL is ongoing.

Introduction: The treatment of chronic lymphocytic leukemia (CLL) has evolved over the past two decades, from finite duration chemoimmunotherapy (CIT) to an era of targeted agents such as Bruton Tyrosine Kinase (BTK) inhibitors, BCL2 inhibitors, and anti-CD20 monoclonal antibodies. Little is known about how initial management strategy (watch and wait [W&W] vs. treatment) at diagnosis impacts patient reported outcomes (PROs) in CLL survivors over time, particularly in light of therapeutic advances over the last decade. Methods: Newly diagnosed CLL patients were prospectively enrolled within 9 months of diagnosis in the University of Iowa/Mayo Clinic SPORE Molecular Epidemiology Resource (MER1) study between 9/1/02 and 6/30/15, and within 6 months of diagnosis in the MER Phase 2 (MER2) study between 7/1/15 and 5/1/20. PROs were measured using the Functional Assessment of Cancer Therapy-General scale (FACT-G), which generates Physical (PWB, range 0-28), Social/Family (SFWB, 0-28), Emotional (EWB, 0-24), and Functional Well-Being (FWB, 0-28) scores and a Total Score (0-108), with higher scores indicating better WB. In both cohorts, FACT-G was measured at enrollment and at years 1, 2, and 3 after diagnosis, while the timing of later assessments varied slightly by study (MER1: 6 yrs and 9 yrs; MER2: 5 years). We employed a generalized linear mixed model to evaluate change from baseline between groups with groups defined by each participant's initial management strategy (W&W vs. upfront systemic treatment). Initial management strategy was determined by the treating physician. The study was approved by the Mayo Clinic IRB. Results: A total of 722 patients were identified in MER1 and 256 patients in MER2. The median age at diagnosis was 63 years in MER1 and 66 years in MER2 (P<0.05); 66% were male in MER1 and 68% were male in MER2 (P>0.05). More patients initiated upfront systemic therapy in MER1 (n=380, 53%) than in MER2 (n=110, 43%). Other baseline clinical characteristics were similar between the cohorts ( Table 1). There was no statistical difference in the mean FACT-G total score at baseline between MER1 (85.0, SD 12.5) and MER2 (85.2, SD 12.7). In the MER1 cohort, statistically significant lower baseline QOL scores were observed in patients initiating upfront treatment compared to W&W on the FACT-G total score (P=0.002) as well as the PWB (P=0.01), EWB (P=0.03), and FWB (P=0.006) subscales. We also observed that over time, EWB and FWB scores were consistently lower amongst those treated upfront compared to W&W. Further, EWB scores steadily improved over time regardless of initial management ( Figure 1), whereas the FWB and PWB subscales were largely stable over time. Mean SFWB scores dropped amongst patients who received upfront therapy at 3 years, which then steadily improved through year 9 but did not recover back to baseline. SFWB scores were stable among W&W patients. In the MER2 cohort, no statistically significant differences in any of the baseline QOL scores were noted between the W&W vs. initial treatment groups. Despite this, we observed consistently lower mean EWB scores among patients who initiated upfront treatment compared to W&W pts over time. No other statistically significant differences in mean QOL scores over time were observed between groups. Similar to MER1, we did observe a drop in mean SFWB scores amongst patients initiating upfront therapy at 3 years of follow-up that remained decreased at year 5. Conclusion: This is the largest longitudinal study of CLL survivors that spanned over 18 years and two treatment eras. We demonstrated that in the novel treatment era, QOL metrics were similar between initial management strategies (W&W vs. treatment), suggesting that initial management had little to no impact on QOL at baseline or over time. In contrast, in the CIT era, we observed differences in QOL metrics at baseline between groups. In both treatment eras, patients with CLL who required upfront therapy demonstrated a decrease in SFWB from baseline at year 3 of follow-up, and the mean EWB scores were lowest at baseline regardless of initial management strategy (W&W vs. upfront treatment) or treatment era (CIT vs. novel agents), but steadily improved over time. These findings suggest opportunities for interventions to improve QOL in patients with CLL, and additional analyses are ongoing. Further follow-up in the era of novel agents is needed.

Multiple Myeloma (MM) progresses from monoclonal gammopathy of undetermined significance (MGUS) and/or smoldering MM (SMM). Despite similar primary genetic events as MM, MGUS and SMM PCs are non- or low-proliferative. Since oncogene activation and DNA damage drive senescence growth arrest to promote immune clearance of potentially malignant cells, we hypothesized that MGUS and SMM pre-malignant PCs exhibit senescence features. To test this, we performed gene set enrichment analysis (GSEA) of a published human PC gene array dataset (GSE5900) comparing normal, MGUS, and SMM PCs for custom senescence phenotyping gene sets (Senescence Up, SenUp; Senescent Cell Anti-Apoptosis Pathways, SCAPs; Senescence Growth Arrest, SenGA; Inflammatory Senescence Associated Secretory Phenotype, iSASP; Interferon SASP, IFN-SASP) versus biological aging gene sets (SenMayo; CellAge). MGUS and SMM PCs exhibited significant enrichment (NES>1.5, q<0.05) for senescence gene sets SenUp, SenGA, and SCAPs compared to normal PCs, with no enrichment for aging gene sets. MGUS PCs were also enriched for iSASP (NES=1.4, q=0.11), whereas SMM PCs were enriched for IFN-SASP (NES=2.3, q<0.05), characteristic of late-senescence. Thus, both MGUS and SMM PCs show enrichment of senescence gene sets compared to normal PCs with differential SASP profiles. Next, we evaluated MGUS, SMM, and MM patient PCs isolated by magnetic sorting for histological senescence features. PCs were immunostained for LMNB1 and HMGB1, and fluorescent in situ hybridized for α-satellite. Cellular senescence was defined as loss of ≥50% nuclear membrane LMNB1, loss of HMGB1, and ≥2 senescence-associated distensions of satellite (SADS) per nuclei. PCs from MGUS patients (N=12) exhibited increased percentage of senescent PCs (26.9±4.1%, Mean±SEM, p<0.0001) compared to SMM (11.0±1.9%, N=19, p<0.0001) and MM (11.6±1.5%, N=11, p<0.001). A subset analysis of SMM patients for which disease stability was known (≥2 years, N=6-8) showed no differences. These results demonstrate that the percentage of senescent PCs is increased in MGUS but not SMM PCs compared to proliferative disease. To assess PC senescence in the native bone marrow microenvironment (BMME), we evaluated trephine bone biopsies from MGUS and SMM patients that progressed or not to MM ≤10 years, as well as newly diagnosed MM. Biopsies were immunostained for CD138, LMNB1, and HMGB1. Whole biopsy images were analyzed using the artificial intelligence-based software HALO. Cells were classified as senescent if they were double negative and non-senescent if they were double positive for LMNB1 and HMGB1. The percentage of non-senescent PCs increased with progression to MM and positively correlated with total PC burden (R=0.52, p<0.0001), reflective of proliferative disease. Senescent PCs correlated with total senescent BMME (R=0.87, p<0.0001), consistent with a role for senescent cells to drive paracrine senescence. We performed proximity analysis to evaluate the cellular composition within 25um of senescent and non-senescent PCs. As expected, non-senescent PCs exhibited a significantly greater percentage of proximate non-senescent BMME in all groups (p<0.0001). In contrast, senescent PCs in stable MGUS (N=16) and SMM (N=14) were equally surrounded by senescent and non-senescent BMME. In progressing MGUS (N=21) and SMM (N=20), as well as MM (N=18), senescent PCs exhibited significantly reduced senescent BMME in their proximate environment (p<0.05). This suggests that senescent PCs in progressive disease, in contrast to stable disease, fail to drive local paracrine senescence responses. These data demonstrate that PCs from stable MGUS and SMM patients exhibit senescence features, with isolated PCs from stable MGUS showing increased senescence gene expression and histologic features. While stable SMM also exhibits senescence gene expression, they lose morphological features of senescence and show a preferential IFN SASP. Within the bone compartment, senescent PCs in stable MGUS and SMM exhibit the expected proximity to senescent BMME. In contrast, progressing MGUS and SMM, as well as NDMM, show a loss of proximate senescent BMME. Given the role of senescence to drive inflammation to clear potentially tumorigenic cells, these data support that failure to drive paracrine senescence may be key to the immune escape of pre-malignant cells in MGUS and SMM.

INTRODUCTION Telomere biology disorders (TBDs) are multisystem diseases caused by accelerated telomere shortening that can present with bone marrow failure and pulmonary fibrosis, among other abnormalities. TBD diagnosis is considered when average blood telomere length (TL) measured by clinical flowFISH is <1 st percentile and/or a pathogenic variant in telomere maintenance genes (most commonly TERT, TERC, RTEL1) is identified. Pathogenic variants in genes coding for the telomerase protein ( TERT) and RNA component ( TERC) decrease telomerase activity. Similarly, variants in RTEL1 result in defective unwinding of the t-loop at telomere ends, increasing the levels of free t-circles in cells. Variants in these genes are often classified as variants of uncertain significance (VUSs by ACMG criteria) with additional efforts needed for variant curation. To address this, we describe results of functional testing performed in TBD patients with VUSs involving TERT, TERC, and RTEL1. METHODS TBD suspected patients were identified through the Pre-Myeloid Cancer and Bone Marrow Failure Clinic (Mayo Clinic) where they underwent clinical flowFISH and genetic testing for TBD related genes. Informed consent was obtained with approval from the Mayo Clinic Institutional Review Board. Viable peripheral blood and bone marrow samples were collected along with age matched healthy volunteers. Telomerase activity was measured in samples with TERT and TERC variants through telomeric repeat amplification protocol (TRAP) assays using the TRAPeze XL Telomerase Detection Kit (MilliporeSigma) following manufacturer instructions. HCT116 cells were used as positive controls while heat inactivated samples served as negative controls. Fluorescence was measured in a SpectraMax plate reader. Percentage of total product generated (TPG) relative to healthy age matched controls was calculated. T-circle detection assays were used on RTEL1 samples based on the protocol by Zhang et al. (2017). Final southern blotting was completed using the TeloTAGGG Telomere Length Assay Kit (Roche), and t-circle bands were visualized in a c600 GEL Imaging System (Azure Biosystems). RESULTS Ten patients (mean age = 53 years (27-71), 50% female) were included in our study. Patients 1 to 5 presented VUS in TERT (n=4) or TERC (n=1) and TL <1 st percentile in lymphocytes and/or granulocytes. Patient 6 carried a known pathogenic variant in RTEL1 and was included as a positive control. Patients 7 to 9 with RTEL1 variants presented TL close to the 10 th or between the 1 st-10 th percentile. No TL results were available for patient 10 but was tested due to the presence of a VUS in RTEL1 (Table 1). Similarly, while patient 7 carried a likely pathogenic variant, it was tested due to presenting lymphocyte TL between the 1 st-10 th percentile. TRAP results indicated decreased telomerase activity compared to age matched controls in all samples with TERT and TERC variants suggesting a pathogenic effect (52.%, 10.7%, 48.7%, 4.4% and 5.2% of TPG compared to age matched controls, respectively) (Table 1). On the other hand, t-circles were only seen in patient 6 (Figure 2). DISCUSSION We used TRAP and t-circle assays to explore the telomerase activity and presence of t-circles in patient samples with TERT/TERC and RTEL1 variants. In the case of TERT/TERC, we observed a reduction of telomerase activity in all samples. Together with blood TL below the 1 st percentile, this suggests that these variants are indeed deleterious (Table 1). However, since patient 4 presented two TERT VUSs, additional testing is needed to conclude how each variant impacts telomerase activity independently. In regard to RTEL1 variants, we observed t-circles in patient 6 as expected. No t-circles could be seen on all remaining samples, including patient 7 who carried a variant classified as likely pathogenic (Figure 1). This in-frame deletion has been described once in a TBD family (Cogan et al. 2015) but was not functionally tested. This result, together with the borderline TL seen in this patient, may suggest that some compensatory mechanism exists on this sample, or that the deleterious effect of the variant is not strong enough to be detected by our assay. Further studies are currently ongoing by our group to validate these results using a cell-based transfection disease model. In summary, we show that functional testing can be a helpful tool to determine pathogenicity of VUSs in TERT/TERC and RTEL1.

Introduction: In many cases, treatment decisions for multiple myeloma patients must be made in the absence of high quality randomized controlled clinical trials. As a result, many clinicians lean on prognostic models to drive treatment selection. The most widely used of these models incorporate a limited number of patient related factors that reflect tumor burden, cytogenetic features, or gene expression profiling. These prediction scores are derived from regression models that incorporate these variables at the time of diagnosis. They do not account for variables that change dynamically over time, do not explore the relationship between treatment selection and progression, and imperfectly predict survival outcomes. We designed a prognostic model based on an ensemble machine learning platform to predict progression in NDMM using Extreme Gradient Boosting Machine (XGBoost) combined with accelerated failure time modeling for survival analysis. Methods: We utilized a large retrospective data set containing treatment and response information for 1127 patients with newly diagnosed multiple myeloma (NDMM) treated at the Cleveland Clinic Foundation between 2000 and 2023. Following data preprocessing based on data completeness, 953 patient records were included in our training, testing, and cross validation datasets. For the purposes of our analysis, the initial dataset was randomly split into training (70%), testing (15%) and validation (15%) subsets at the patient level. We also defined the lower and upper bounds for each subset, which is critical due to the right-censored nature of survival data. Hyperparameters for XGBoost were optimized using Bayesian search, minimizing the negative log-likelihood. A random forest method was applied for the imputation of missing data (MissForest). We then trained the XGBoost model using GPU acceleration for enhanced computational efficiency. Finally, a log-rank test on predicted survival times was used to test the model's performance on patients with and without known progression. Results: Our preprocessed data set had 953 patients with a mean age at disease onset of 65 years and a slight male predominance (55%). Approximately 27% of patients harbored high cytogenetic risk disease and 3.5% of the cohort presented with extramedullary disease a diagnosis. At a median follow up of 35 months, 47% of patients had experienced disease progression or death. Induction therapy included an immunomodulatory drug in 40% of patients and proteasome inhibitors in 22%. Frontline autologous stem cell transplantation followed induction in 28% of patients. Median progression free survival in the overall data set was 44 months. These features were consistent across the randomly assigned training, testing, and validation cohorts. Following data preprocessing, 34 independent clinical and genomic variables including selection of first line treatment were assessed as model inputs each available record. Under optimized parameters, the model was trained with a maximum tree depth = 7 and a learning rate ~ 0.22 for the desired output of progression free survival (PFS). We then queried PFS for both the training and validation sets and divided patients into those known to have progressed and those without a progression event. Survival analysis was undertaken using predicted PFS values for patients known to have progressed and those who remained progression free. For the validation data set, our model successfully discerned between progressed and non-progressed cases (log-rank test statistic = 22.07, p < 0.005) (Figure 1). Discussion: We present a machine learning approach based on regularized gradient boosting that accurately discerns between patients who experience progression and those who remain progression free at a median ~35 months of follow-up in a large retrospective data set in patients presenting with NDMM. Further elaboration of our model will allow for the incorporation of large amounts data to predict survival outcomes on the basis of dynamic variables such as depth of response and treatment selection. With the ability to parse outcomes among multiple myeloma patients at high resolution, future clinicians and clinical trialists may be able to overcome limitations in trial design and patient-level therapy selection for both newly diagnosed and relapsed patients.

Introduction Fluorescent In-Situ Hybridization (FISH) has been an indispensable tool to characterize the cytogenetic landscape of multiple myeloma (MM), mainly for identifying primary abnormalities such as heavy chain gene (IgH) translocations and odd-number chromosome trisomies. These aberrations play a crucial role in risk stratifying the disease given their correlation with disease aggressiveness, treatment response and overall prognosis. High-risk (HR) cytogenetics are identified through the presence of t(14;16), t(14;20) or loss of the p53 gene locus. Less is known about the prognostic implication of secondary abnormalities, in particular gain of 1q21. We aimed to evaluate the association between primary and secondary cytogenetic abnormalities and their impact on overall survival (OS) in MM patients. Methods We performed a retrospective analysis on a FISH records database from three Mayo Clinic campuses. Cohort included patients >18 years of age, diagnosed with multiple myeloma between 2005 and 2022, with FISH performed within two years of diagnosis. Demographics including age, date of diagnoses and OS data were retrieved from electronic medical records. Statistical analyses were done using BlueSky Statistics software. Study was conducted in accordance with the principles of the Declaration of Helsinki; informed consent was obtained from all participants. Results We analyzed a total of 2221 patients, among which 59.8% were male, with a median age of diagnosis 65 years. Hyperdiploidy was the most prevalent primary cytogenetic abnormality observed in 44.3% of patients, followed by t(11;14) as the most common translocation (23%), and t(4;14) (8.2%). Gain(1q21) was assessed in 925 patients, among which 64%, 30.3% and 5.7% had 2, 3 and ≥4 copies respectively. Furthermore, 11.4% (n=253) exhibited either mutated TP53 or del17p, and 37.8% (n=839) presented with 13q loss or monosomy 13. The median follow-up for the overall population was 102 months (8.5 years), with a median OS of 75.5 months. Patients with high-risk translocations had significantly shorter median OS of 40.01 months compared to 80.65 months for those with standard risk translocations (p <0.001). The presence of ≥4 1q copies was associated with significantly reduced overall survival (2 copies: not reached, 3 copies: 69.4 months, ≥4 copies: 40.1774; p <0.001). This was consistent even when considering the presence of high-risk translocations (Figure 1). To confirm the independent significance of these findings, we performed multivariate Cox regression models, incorporating age at diagnosis, year of diagnosis, gender, presence of hyperdiploid disease, and presence of high-risk translocations. Given that overall survival outcomes were analyzed irrespective of treatment received, we incorporated year of diagnosis as a variable to control for potential impact of evolving treatment regimens over the years. The presence of ≥4 1q copies was associated with reduced OS, with a hazard ratio (HR) of 2.12 (CI: 1.38-3.26; p <0.001). The presence of 3 copies also showed prognostic significance (HR: 1.43; p: 0.012). Similarly, both the presence of TP53mut/mon17 (HR: 1.96, CI: 1.66-2.31; p <0.001) and 13qdel/mon13 (HR: 1.21, CI: 1.08-1.38) were associated with decreased OS. On the other hand, 1p deletion did not have any significant prognostic implications (HR: 1.84, CI: 0.38-1.59; p 0.113). When incorporating all cytogenetic events into a multivariate model, variables associated with reduced OS included age at diagnosis, the presence of both of 3 and ≥4 1q copies, TP53mut/mon17, t(4;14), t(14;16) and t(14;20). However, 13qDel/mon13 lost significance when incorporated with other abnormalities (Table 1). Conclusion Our investigation sheds light on the less explored realm of secondary cytogenetic abnormalities and confirm a strong association between increasing number of 1q copies and reduced overall survival, independent of other cytogenetic abnormalities and treatment regimens. These results underscore the utility of FISH-based comprehensive cytogenetic profiling at baseline, which enables reliable identification of high-risk patients. Future prospective trials should further elucidate the pathogenic mechanism for these findings. Our analysis used OS as primary endpoint and included over 2000 MM patients, reinforcing the clinical relevance of our findings.