Acute myeloid leukemia (AML) is a complex hematological malignancy with multiple disease sub-groups defined by somatic mutations and heterogeneous outcomes. Although genome-wide association studies (GWAS) have identified a small number of common genetic variants influencing AML risk, the heritable component of this disease outside of familial susceptibility remains largely undefined. Here we perform a meta-analysis of four published GWAS plus two new GWAS, totalling 4710 AML cases and 12938 controls. We identify a new genome-wide significant risk locus for pan-AML at 2p23.3 (rs4665765; P=1.35x10-8; EFR3B, POMC, DNMT3A, DNAJC27) which also significantly associates with patient survival (P=6.09x10-3). Our analysis also identifies three new genome-wide significant risk loci for disease sub-groups, including AML with deletions of chromosome 5 and/or 7 at 1q23.3 (rs12078864; P=7.0x10-10; DUSP23) and cytogenetically complex AML at 2q33.3 (rs12988876; P=3.28x10-8; PARD3B) and 2p21 (rs79918355; P=1.60x10-9; EPCAM). We also investigated loci previously associated with risk of clonal hematopoiesis (CH) or clonal hematopoiesis of indeterminate potential (CHIP) and identified several variants associated with risk of AML. Our results further inform on AML etiology and demonstrate the existence of disease sub-group specific risk loci.

BACKGROUND:Initial results of this study, reported after a median follow-up close to 4 years, demonstrated improved time to initiation of new treatment (TTNT) for patients with advanced stage, asymptomatic, low tumour burden follicular lymphoma who received early rituximab monotherapy when compared with watchful waiting. Given the long natural history of follicular lymphoma, the trial was extended to further assess TTNT with longer follow-up. Mature data are presented here. METHODS:In this open-label, randomised, phase 3 trial, conducted at 118 centres in five countries, adult patients with asymptomatic, stage II-IV, grade 1-3a low tumour burden follicular lymphoma and Eastern Cooperative Oncology Group performance status 0-1 were randomly assigned (1:1:1) between watchful waiting, rituximab induction (375 mg/m2, intravenous) weekly for four doses (rituximab induction group) and rituximab induction followed by rituximab maintenance at the same dose every 8 weeks for 12 doses (rituximab maintenance group). The rituximab induction group closed early on Sept 30, 2007, and the study was amended to a two-arm trial. The primary endpoint was TTNT, assessed in the intention-to-treat population. The study is registered with ClinicalTrials.gov, NCT00112931, and recruitment and follow-up are complete. FINDINGS:Between Oct 15, 2004, and May 1, 2009, 455 patients were randomly assigned, including 183 to watchful waiting, 82 to rituximab induction, and 190 to rituximab maintenance. Median follow-up was 14·7 years (IQR 13·3-15·6). At 15 years, 65% (95% CI 56-72) of patients in the rituximab maintenance group, 48% (36-60) in the rituximab induction group, and 34% (27-42) in the watchful waiting group had not started new treatment. Median TTNT was not yet reached (95% CI 15·6-not estimable) in the rituximab maintenance group, 14·8 years (7·5-not reached) in the rituximab induction group, and 5·6 years (3·8-8·4) in the watchful waiting group. TTNT was longer in both the rituximab induction and rituximab maintenance groups compared with the watchful waiting group (rituximab induction vs watchful waiting: hazard ratio [HR] 0·55 [95% CI 0·38-0·80], p=0·0019; rituximab maintenance vs watchful waiting: HR 0·36 [0·26-0·50], p<0·0001). INTERPRETATION:These mature data with 15 years of follow-up confirm that early rituximab monotherapy substantially delays the need for new treatment for patients with advanced stage, asymptomatic low tumour burden follicular lymphoma, providing an evidence base for its use in this setting and confirming its value for patients who seek to defer or avoid treatment with chemotherapy. FUNDING:Cancer Research UK, Lymphoma Research Trust, Lymphoma Association, and Roche.

Despite recent refinements in the diagnostic and prognostic assessment of CEBPA mutations in AML, several questions remain open, i.e. implications of different types of basic region leucin zipper (bZIP) mutations, the role of co-mutations and the allelic state. Using pooled primary data analysis on 1010 CEBPA-mutant adult AML patients, a comparison was performed taking into account the type of mutation (bZIP: either typical in-frame insertion/deletion (InDel) mutations (bZIPInDel), frameshift InDel or nonsense mutations inducing translational stop (bZIPSTOP) or single base-pair missense alterations (bZIPms), and transcription activation domain (TAD) mutations) and the allelic state (single (smCEBPA) vs. double mutant (dmCEBPA)). Only bZIPInDel patients had significantly higher rates of complete remission and longer relapse free and overall survival (OS) compared with all other CEBPA-mutant subgroups. Moreover, co-mutations in bZIPInDel patients (e.g. GATA2, FLT3, WT1 as well as ELN2022 adverse risk aberrations) had no independent impact on OS, whereas in non-bZIPInDel patients, grouping according to ELN2022 recommendations added significant prognostic information. In conclusion, these results demonstrate bZIPInDel mutations to be the major independent determinant of outcome in CEBPA-mutant AML, thereby refining current classifications according to WHO (including all dmCEBPA and smCEBPA bZIP) as well as ELN2022 and ICC recommendations (including CEBPA bZIPms).

Abstract Background Deletions and partial losses of chromosome 7 (chr7) are frequent in acute myeloid leukemia (AML) and are linked to dismal outcome. However, the genomic landscape and prognostic impact of concomitant genetic aberrations remain incompletely understood. Methods To discover genetic lesions in adult AML patients with aberrations of chromosome 7 [abn(7)], 60 paired diagnostic/remission samples were investigated by whole-exome sequencing in the exploration cohort. Subsequently, a gene panel including 66 genes and a SNP backbone for copy-number variation detection was designed and applied to the remaining samples of the validation cohort. In total, 519 patients were investigated, of which 415 received intensive induction treatment, typically containing a combination of cytarabine and anthracyclines. Results In the exploration cohort, the most frequently mutated gene was TP53 (33%), followed by epigenetic regulators (DNMT3A, KMT2C, IDH2) and signaling genes (NRAS, PTPN11). Thirty percent of 519 patients harbored ≥ 1 mutation in genes located in commonly deleted regions of chr7—most frequently affecting KMT2C (16%) and EZH2 (10%). KMT2C mutations were often subclonal and enriched in patients with del(7q), de novo or core-binding factor AML (45%). Cancer cell fraction analysis and reconstruction of mutation acquisition identified TP53 mutations as mainly disease-initiating events, while del(7q) or −7 appeared as subclonal events in one-third of cases. Multivariable analysis identified five genetic lesions with significant prognostic impact in intensively treated AML patients with abn(7). Mutations in TP53 and PTPN11 (11%) showed the strongest association with worse overall survival (OS, TP53: hazard ratio [HR], 2.53 [95% CI 1.66–3.86]; P < 0.001; PTPN11: HR, 2.24 [95% CI 1.56–3.22]; P < 0.001) and relapse-free survival (RFS, TP53: HR, 2.3 [95% CI 1.25–4.26]; P = 0.008; PTPN11: HR, 2.32 [95% CI 1.33–4.04]; P = 0.003). By contrast, IDH2-mutated patients (9%) displayed prolonged OS (HR, 0.51 [95% CI 0.30–0.88]; P = 0.0015) and durable responses (RFS: HR, 0.5 [95% CI 0.26–0.96]; P = 0.036). Conclusion This work unraveled formerly underestimated genetic lesions and provides a comprehensive overview of the spectrum of recurrent gene mutations and their clinical relevance in AML with abn(7). KMT2C mutations are among the most frequent gene mutations in this heterogeneous AML subgroup and warrant further functional investigation.

Relapse following CD19-directed chimeric antigen receptor T-cells (CAR-T) for relapsed/refractory large B-cell lymphoma (r/r LBCL) is commonly ascribed to antigen loss or CAR-T exhaustion. Multi-antigen targeting and PD-1 blockade are rational approaches to prevent relapse. Here, we test CD19/22 dual-targeting CAR-T (AUTO3) plus pembrolizumab in r/r LBCL as inpatient or outpatient therapy (NCT03289455, https://clinicaltrials.gov/ct2/show/NCT03289455). Endpoints include toxicity (primary) and response rates (secondary). AUTO3 was manufactured for 62 patients using autologous leukapheresis, modified with a bicistronic transgene. 52 patients received AUTO3 (7/52,50x106; 45/52,150-450x106) and 48/52 received pembrolizumab. Median age was 59 years (range,27-83) and 46/52 had stage III/IV disease. Median follow-up was 21.6 months (range,15.1-51.3) at last data cut (Feb 28, 2022). AUTO3 was safe: grade 1-2 and grade 3 CRS affected 18/52 (34.6%) and 1/52 (1.9%) patients, neurotoxicity arose in 4 patients (2/4, grade 3-4), HLH affected 2 patients, and no Pembrolizumab-associated autoimmune sequalae were observed. On this basis, outpatient administration was tested in 20 patients, saving a median of 14 hospital days/patient. AUTO3 was effective: overall response rates were 66% (48.9%, CR; 17%, PR). For patients with CR, median DOR was not reached, with 54.4% (CI: 32.8, 71.7) projected to remain progression-free beyond 12 months after onset of remission. DOR for all responding patients was 8.3 months (95% CI: 3.0, NE) with 42.6% projected to remain progression-free beyond 12 months after onset of remission. Overall, AUTO3 +/- pembrolizumab for r/r LBCL was safe, lending itself to outpatient administration, and delivered durable remissions in 54.4% of complete responders, associated with robust CAR-T expansion. Neither dual-targeting CAR-T nor pembrolizumab prevented relapse in a significant proportion of patients, and future developments include next-generation-AUTO3, engineered for superior expansion/persistence in vivo, and selection of CAR binders active at low antigen densities.

Background: Obe-cel is an autologous CD19 chimeric antigen receptor (CAR) T cell product designed to reduce toxicity and improve persistence through a fast off-rate CD19 binding domain. The clinical activity of obe-cel has been explored in adults with R/R B-ALL in a Phase I study (ALLCAR19, NCT02935257; Roddie C et al. J Clin Oncol 2021) and a Phase Ib/II study (FELIX, NCT04404660; Roddie C et al. J Clin Oncol 2023;41[16 Suppl]:7000). Additionally, obe-cel has been tested in pts with R/R B-cell chronic lymphocytic leukemia (B-CLL) and R/R B-cell non-Hodgkin lymphoma (B-NHL) (ALLCAR19 extension; Roddie C et al. Blood 2022;140[1 Suppl]:7452-3). Pts from the ALLCAR19 and FELIX Phase Ib studies are in long-term follow up (≥22 mos), and the ALLCAR19 extension has been recruiting for 3 years. We report an analysis of long-term efficacy and safety data from the ALLCAR19 and FELIX Phase Ib studies, as well as data from the ALLCAR19 extension. Methods: ALLCAR19 is a multicenter, non-randomized, open-label Phase I study in pts aged ≥16 years with B-cell malignancies. ALLCAR19 initially recruited pts with R/R B-ALL but was then amended (extension study) to also include pts with R/R B-CLL and R/R B-NHL. FELIX is a global, single-arm Phase Ib/II study enrolling pts aged ≥18 years with R/R B-ALL.Study designs have been presented previously. Obe-cel was administered as a split dose in pts with B-ALL (target dose 410 × 10 6 CAR T cells) and pts with CLL (target dose 230 × 10 6 CAR T cells), and as a single infusion in pts with B-NHL (target dose 200 × 10 6 CAR T cells); the pt populations in the two studies were similar. Pts with B-ALL from the ALLCAR19 and FELIX Phase Ib studies are being followed long term for disease progression and survival. For this analysis, data in pts with B-ALL from the ALLCAR19 and FELIX Phase Ib studies were pooled. Data in pts with CLL or B-NHL are presented from the ALLCAR19 extension study. Results: Outcomes in pts with R/R B-ALL: Data in pts with B-ALL were pooled (20 pts from ALLCAR19 [data cut-off Jun 26, 2023] and 16 from FELIX Phase Ib [data cut-off Mar 16, 2023]). The median age of the pooled cohort was 41.5 (range 18 to 74) years and pts had received a median of 3 (range 2 to 6) prior lines of treatment. Twenty-nine of the 36 pts (81%) achieved complete remission (CR)/CR with incomplete hematologic recovery post obe-cel infusions, per investigator assessment. The event-free survival rate was 64% at 6 mos and 49% at 12 mos. With a median follow up of 43 (range 19 to 62) mos, 13/36 pts (36%) remain in remission (8 from ALLCAR19; 5 from FELIX Phase Ib). Among these 13 ongoing responders, 2 (15%) had consolidation with allogeneic hematopoietic stem cell transplantation (allo-HSCT). Ten of the 11 ongoing responders (91%) who did not receive allo-HSCT still had detectable CAR T cells at the last follow up. All ongoing remissions were measurable residual disease negative at last available assessment. The estimated 2-, 3- and 4-year overall survival rates were 44%, 39% and 39%, respectively. Outcomes in pts with R/R B-CLL/B-NHL: The extension phase of the ALLCAR19 study enrolled 35 pts with B-CLL or B-NHL, of which 26 (B-CLL n=5; B-NHL n=21) received obe-cel (data cut-off Jun 26, 2023). The median age of this combined cohort was 61 (range 39 to 79) years and pts had received a median of 3 (range 2 to 8) prior lines of treatment. At a median follow up of 24 mos, the overall response rate for this cohort was 92% (n=24), and 58% of responders (n=14) were alive without disease progression at last follow up. Late toxicity: Of the 11 long-term R/R B-ALL responders who had not received consolidation allo-HSCT, 10 have ongoing B-cell aplasia. Of the 14 ongoing responders in the R/R B-CLL/B-NHL cohort, 12 have ongoing B-cell aplasia (<20 B cells/µl). Of note, ongoing B-cell aplasia did not correlate with an increased risk of late serious infection. No other long-term toxicity ascribed to obe-cel was reported. Conclusions: The combined analysis of data from the ALLCAR19 and FELIX Phase Ib studies shows long-term efficacy and safety of obe-cel in pts with R/R B-ALL, with approximately one-third of pts still in remission without consolidative allo-HSCT after a median follow up of >3 years. Durable responses of >2 years were also seen in pts with R/R B-CLL and R/R B-NHL.B-cell aplasia was commonly found in long-term follow up of obe-cel recipients, but without a corresponding rise in serious late infections. Obe-cel can effect durable long-term remissions in B-cell malignancies.

Background: In adults with acute myeloid leukaemia (AML), certain genetic abnormalities, such as a complex karyotype and/or mutations in TP53, are well-recognized to be associated with failure to achieve remission after induction chemotherapy, so-called induction failure (IF). Despite IF occurring up to 10% of normal karyotype (NK-AML), the genetic and transcriptional basis of chemotherapy resistance in this subgroup is less well characterized (1). Previous studies on small patient groups did not identify clear associations of IF with genetic mutations in NK-AML, emphasizing the need for a more comprehensive genomic analysis utilizing a larger sample size (2). Aims: To identify genomic driver mutations and transcriptomic signatures in NK-AML IF through analysis of primary patient samples. Methods: DNA samples from 70 NK-AML patients who experienced IF, and 70 matched NK-AML patients who achieved remission, were sequenced using the All Exon V6+UTR capture probes (Agilent) using the NovaSeq platform to reach a mean depth of 100X. Variant calling was done by the GATK 4.1.9 bundle. Paired-end mRNA sequencing was performed at 30 million reads using Illumina Novaseq platform on 43 NK-AML IF samples vs 48 chemoresponsive samples (n=48) matched for karyotype, age, sex, sample source, NPM1 and FLT3 mutational status. RNAseq analysis was performed by Nextflow nf-core RNAseq pipeline. Results: In IF NK-AML, the top mutated genes from WES were DNMT3A (39%), FLT3-ITD (33%) and NPM1 (29%; Fig 1). Most strikingly, WT1 mutations were present in 29% of IF patients and had a high mutual co-exclusivity with DNMT3A mutations (p<0.05), consistent with their opposing roles in DNA methylation. TP53 mutations were surprisingly rare, present in only 2 patients with IF. Unexpectedly, APC mutations occurred in 15.7% of IF patients. Gene-set enrichment analysis yielded a significant senescence signature in IF patients from two different datasets in The Broads Institute’s Molecular Signatures Database v7.5.1 (NES>2, FDR<0.01). This was supported by the enrichment of the accompanying senescence-associated secretory phenotype (SASP; NES=2.7, FDR<0.01) and quiescence signature in the IF group (NES=2.2, FDR<0.01). While previous studies have implicated chemotherapy-induced senescence as a mechanism of chemotherapy resistance (3), our study suggests a senescence phenotype can occur prior to chemotherapy leading to IF in a subset of patients. We speculate this is genetically associated with the high incidence of WT1 and APC mutations in this group of patients, given both genes have been implicated in regulating cellular senescence in other disease models (4-5). Furthermore, this study identifies a genetic and transcriptional signature of patients with NK-AML likely to experience IF that might prove a useful biomarker if confirmed on independent cohorts. Summary/Conclusion: • WES analysis in NK-AML IF identified a high incidence of WT1 mutations that are mutually exclusive with DNMT3A mutations. • APC mutations were unexpectedly enriched in NK-AML IF patients. • RNAseq revealed the enrichment of senescence and quiescence in NK-AML IF that proceeds chemotherapy. References: 1. Gaidzik V. Semin Oncol. 2008;35(4):346-55. 2. Brown FC. Br J Haematol. 2017;176(1):86-91. 3. Duy C. Cancer Discov. 2021;11(6):1542-61. 4. Hernandez-Segura A. Trends Cell Biol. 2018;28(6):436-53. 5. Vicent S. J Clin Invest. 2010;120(11):3940-52.Keywords: Senescence, Genomics, Chemoresistance, AML

Supplementary Table 2: Cell viability data from the small-molecule drug screening in EZH2-WT and EZH2-KO1 cells (related to Fig. 1)

Background:The clinicopathological spectrum of nodular lymphocyte predominant Hodgkin lymphoma (NLPHL), also known as nodular lymphocyte predominant B-cell lymphoma, partially overlaps with T-cell/histiocyte-rich large B-cell lymphoma (THRLCBL). NLPHL histology may vary in architecture and B-cell/T-cell composition of the tumour microenvironment. However, the immune cell phenotypes accompanying different histological patterns remain poorly characterised.Methods:We applied a multiplexed immunofluorescence workflow to identify differential expansion/depletion of multiple microenvironmental immune cell phenotypes between cases of NLPHL showing different histological patterns (as described by Fan et al, 2003) and cases of THRLBCL.Results:FOXP3-expressing T-regulatory cells were conspicuously depleted across all NLPHL cases. As histology progressed to variant Fan patterns C and E of NLPHL and to THRLBCL, there were progressive expansions of cytotoxic granzyme-B-expressing natural killer and CD8-positive T-cells, PD1-expressing CD8-positive T-cells, and CD163-positive macrophages including a PDL1-expressing subset. These occurred in parallel to depletion of NKG2A-expressing natural killer and CD8-positive T-cells.Discussion:These findings provide new insights on the immunoregulatory mechanisms involved in NLPHL and THLRBCL pathogenesis, and are supportive of an increasingly proposed biological continuum between these two lymphomas. Additionally, the findings may help establish new biomarkers of high-risk disease, which could support a novel therapeutic program of immune checkpoint interruption targeting the PD1:PDL1 and/or NKG2A:HLA-E axes in the management of high-risk NLPHL and THRLBCL.

Multispectral immunofluorescence (M-IF) analysis is used to investigate the cellular landscape of tissue sections and spatial interaction of cells. However, complex makeup of markers in the images hinders the accurate quantification of cell phenotypes. We developed DeepMIF, a new deep learning (DL) based tool with a graphical user interface (GUI) to detect and quantify cell phenotypes on M-IF images, and visualize whole slide image (WSI) and cell phenotypes. To identify cell phenotypes, we detected cells on the deconvoluted images followed by co-expression analysis to classify cells expressing single or multiple markers. We trained, tested and validated our model on >50k expert single-cell annotations from multiple immune panels on 15 samples of follicular lymphoma patients. Our algorithm obtained a cell classification accuracy and area under the curve (AUC) ≥ 0.98 on an independent validation panel. The cell phenotype identification took on average 27.5 min per WSI, and rendering of the WSI took on average 0.07 minutes. DeepMIF is optimized to run on local computers or high-performance clusters independent of the host platform. These suggest that the DeepMIF is an accurate and efficient tool for the analysis and visualization of M-IF images, leading to the identification of novel prognostic cell phenotypes in tumours.

The spatial architecture of the lymphoid tissue in follicular lymphoma (FL) presents unique challenges to studying its immune microenvironment. We investigated the spatial interplay of T cells, macrophages, myeloid cells and natural killer T cells using multispectral immunofluorescence images of diagnostic biopsies of 32 patients. A deep learning‐based image analysis pipeline was tailored to the needs of follicular lymphoma spatial histology research, enabling the identification of different immune cells within and outside neoplastic follicles. We analyzed the density and spatial co‐localization of immune cells in the inter‐follicular and intra‐follicular regions of follicular lymphoma. Low inter‐follicular density of CD8+FOXP3+ cells and co‐localization of CD8+FOXP3+ with CD4+CD8+ cells were significantly associated with relapse (p = 0.0057 and p = 0.0019, respectively) and shorter time to progression after first‐line treatment (Logrank p = 0.0097 and log‐rank p = 0.0093, respectively). A low inter‐follicular density of CD8+FOXP3+ cells is associated with increased risk of relapse independent of follicular lymphoma international prognostic index (FLIPI) (p = 0.038, Hazard ratio (HR) = 0.42 [0.19, 0.95], but not independent of co‐localization of CD8+FOXP3+ with CD4+CD8+ cells (p = 0.43). Co‐localization of CD8+FOXP3+ with CD4+CD8+ cells is predictors of time to relapse independent of the FLIPI score and density of CD8+FOXP3+ cells (p = 0.027, HR = 0.0019 [7.19 × 10−6, 0.49], This suggests a potential role of inter‐follicular CD8+FOXP3+ and CD4+CD8+ cells in the disease progression of FL, warranting further validation on larger patient cohorts.

INTRODUCTION: AUTO1 is a fast off-rate CD19 binding domain CAR, designed to reduce immune toxicity and improve engraftment. Its clinical activity has been tested in r/r paediatric and adult B-ALL (Ghorashian S et al., Nat Med 2019; Roddie C et al., JCO 2021) and now in adult B-NHL and CLL/SLL (NCT02935257). Here we present data from adult B-NHL and CLL cohorts treated with AUTO1, and report on long-term follow-up of adult B-ALL patients post-AUTO1. METHODS:Manufacturing: AUTO1 products were generated using a closed process from non-mobilised patient leukapheresis. Study design: Subjects ≥ 16y received fludarabine (30mg/m2 x3) and cyclophosphamide (60mg/kg x1) conditioning. DLBCL patients additionally received pembrolizumab (200mg) on day -1. In B-ALL, patients received split dose AUTO1 (day 0: ≥20% Bone Marrow (BM) blasts, AUTO1 dose=10 x106; <20% BM blasts, AUTO1 dose=100 x106. At day+9: if no grade 3-5 CRS/ICANS, dose 2 is administered to a total AUTO1 dose of 410 x106. Split dosing is employed in the CLL cohort (day 0 AUTO1 dose= 30 x106; day 9 AUTO1 dose= 200 x106). B-NHL patients receive a single AUTO1 dose of 200 x106. Study endpoints include manufacture feasibility, grade 3-5 toxicity and remission rates at 1 and 3 months. RESULTS:NHL/CLL cohort: as of 27th July 2022, 28 patients were enrolled: 9 with Follicular Lymphoma (FL), 4 with Mantle Cell Lymphoma (MCL), 8 with DLBCL and 7 with CLL. Apheresis/ manufacture was successful in 26 (1x pending; 1x not harvested due to intracerebral haemorrhage). Median age was 61 years (range 39-79), and patients received a median of 3 prior treatment lines (range 2-8). To date, 23 patients have been infused: 7 FL, 3 MCL, 8 DLBCL and 5 CLL. 2 patients died pre-infusion (1x MCL, COVID-19; 1x CLL, intracerebral haemorrhage). Grade 1 CRS was reported in 7/23 and Grade 2 CRS in 7/23 patients. No ICANS was observed. AUTO1 engraftment was demonstrated in 18/18 patients evaluated by qPCR with ongoing persistence in 17/18 patients at last follow-up. The ORR at month 1 in the FL, MCL and DLBCL cohorts was 7/7 (100%), 3/3 (100%) and 7/8 (88%) respectively, and ongoing responses at last follow-up (FU) were observed in 5/7 FL (FU range, 5-21 months), 2/3 MCL (range, 12-18 months), and 7/8 DLBCL (range, 1-12 months). 2 patients died in remission from COVID-19. In the CLL cohort, 4/5 patients achieved flow negative remission in the bone marrow with ongoing response in all 4 at last FU. B-ALL cohort: of the 20 B-ALL patients treated previously, 8/20 (40%) are in ongoing CR at a median FU of 35 months (IQR 24-36) post-AUTO1, with ongoing B-cell aplasia in 7/8 (89%). CONCLUSIONS: AUTO1 has a tolerable safety profile in patients with r/r B-cell cancers despite high disease burden. In the B-ALL cohort, long-term follow-up indicates that 40% of patients continue in remission post-AUTO1. In both indolent and aggressive NHL and in CLL, AUTO1 shows excellent ORR and CAR engraftment/persistence. Additional patients, updated data and longer follow up will be presented. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Although we earlier reported a very poor outcome for younger adult patients with isocitrate dehydrogenase 2 (IDH2)(R172)-mutated acute myeloid leukaemia (AML) entered into UK trials compared to IDH2(WT) and IDH2(R140)-mutated patients, this was not corroborated by a study from the German-Austrian AML Study Group. We have therefore investigated a later cohort of IDH2-mutated patients to identify any changes in outcome and whether this could inform the optimal treatment for IDH2(R172) AML. We found an improved outcome for IDH2(R172)-mutated AML in the later trials and the data suggests that this may be due to the increased use of allogeneic transplantation to consolidate first remission.

INTRODUCTION We have previously described AUTO1, a CD19 CAR with a fast off-rate CD19 binding domain, designed to reduce CAR-T immune toxicity and improve engraftment. Its clinical activity has been tested in r/r paediatric and adult B-ALL. Cumulatively, this data confirms the intended function of the receptor, with low levels of CRS/ICANS and long-term engraftment of CAR T-cells observed in both patient groups. Recently, CAR-T therapy has been explored in indolent lymphomas such as follicular (FL) and mantle cell lymphoma (MCL), but a high incidence of toxicity including Grade 3-4 ICANS has been reported. We have initiated testing of AUTO1 in the setting of indolent and high-grade B-NHL and CLL (NCT02935257). METHODS Manufacturing: CAR T-cell products were generated using a semi-automated closed process from non-mobilised leukapheresate. Study design: Subjects ≥ 16y underwent lymphodepletion with fludarabine (30mg/m 2 x3) and cyclophosphamide (60mg/kg x1) prior to AUTO1 infusion, with the exception of the DLBCL cohort who additionally received a single dose of pembrolizumab (200mg) on day -1 to potentiate CAR-T expansion. AUTO1 dose varies based on the indication. Split dosing of 230 x10^6 CD19 CAR T-cells at day 0 and day 9 is employed in the CLL cohort. A single dose of 200 x10^6 CD19 CAR T-cells is delivered to patients with B-NHL. Study endpoints include feasibility of manufacture, grade 3-5 toxicity and remission rates at 1 and 3 months. RESULTS As of 17th May 2021, we recruited 13 patients: 7 with FL, 4 with MCL, 1 DLBCL and 1 CLL. Apheresis and product manufacture was successful in all 13 patients and 9 patients were infused: 7 with FL and 2 with MCL. Three patients (1 DLBCL, 1 CLL and 1 MCL) were pending infusion at time of data cut-off and 1 patient (MCL) died due to Covid-19 prior to infusion. Patients treated with AUTO1 had a median age of 56 years (range 39-68y), had received a median of 3 prior lines of treatment (range 2-5) and all patients had stage IV disease at screening. Grade 1 CRS was reported in 4/9 and Grade 2 CRS in 1/9. 1/9 developed MAS which resolved with anakinra/dexamethasone. No ICANS was observed on study. Excellent CAR engraftment was observed and 9/9 patients were in CMR by 18FDG PET-CT post-treatment. At a median of 6.1 months (range 4.0-8.1m), 8/9 patients were disease free at last follow-up. One patient died in CMR at month 5.6 of COVID-19. CONCLUSION AUTO1 has a tolerable safety profile in adult patients with r/r B-NHL despite high disease burden. Early data shows 100% complete remission rates and excellent CAR engraftment/expansion. Additional MCL, CLL and DLBCL patients, updated data and longer follow up will be presented. Roddie: Celgene: Consultancy, Speakers Bureau; Novartis: Consultancy; Gilead: Consultancy, Speakers Bureau. Hartley: Astra Zeneca: Ended employment in the past 24 months; ADC Therapeutics: Consultancy, Current equity holder in publicly-traded company, Current holder of stock options in a privately-held company. Farzaneh: Autolus: Consultancy, Current equity holder in publicly-traded company. Lowdell: Autolus: Consultancy, Current equity holder in publicly-traded company. Linch: Autolus: Consultancy, Current equity holder in publicly-traded company. Pule: Autolus: Current Employment, Current equity holder in publicly-traded company. Peggs: Autolus: Consultancy, Current equity holder in publicly-traded company.

Mutations in SAMD9 or SAMD9L are found in patients with familial predisposition to myelodysplastic syndrome (MDS) and acute myeloid leukaemia (AML) associated with both MIRAGE syndrome (Myelodysplasia, Infection, Restriction of growth, Adrenal hypoplasia, Genital phenotypes, and Enteropathy) and ATXPC (ataxia-pancytopenia syndrome) (Li et al., 1981; Narumi et al., 2016; Chen et al., 2016), respectively. The wild-type (WT) function of SAMD9 and SAMD9L in haematopoietic cell lines is antiproliferative. Germline mutations are thought to be gain-of-function (GOF) mutations, exacerbating the antiproliferative effects of the WT gene (Tesi et al., 2017; Buonocore et al., 2017; Davidsson et al., 2018). Functional somatic reversion through either loss of heterozygosity (−7 or 7q-) or uniparental disomy (UPD), or through acquisition of loss of function (LOF) mutations, has been described in affected individuals. Thus far, LOF mutations have been described in cis for SAMD9, but rarely in trans for SAMD9L (Nagata et al., 2018; Wong et al., 2018). Here we present a four-generation pedigree affected by AML and MDS, with novel disease features and a reversion mutation in trans, consistent with germline mosaicism. The proband (III-1) presented in 1986 aged 10 with a history of excessive bruising. She was one of four siblings, two of which were stillborn. She was also noted to have an extensive family history of MDS and AML; specifically, her mother (II-1) had MDS, an aunt (II-4) died of AML aged 11 following MDS, a cousin (III-5) died of AML aged eight and her mother's cousin (II-8) died of leukaemia (Fig 1A). Her haematological parameters at presentation were normal, except for abnormal platelet function tests. In 1998 (aged 22) she developed tri-lineage cytopenias. Bone marrow aspirate (BMA) at this time showed MDS with 7q- in 5/30 metaphases (46,XX,del(7)(q22q33)[5]/46,XX[25]). Additional multisystem abnormalities were identified over her 20-year follow-up; imaging undertaken for urinary tract infections, adenomyosis and an iliofemoral DVT in 2002 (aged 26) revealed duplex left kidney and ureters, intermediate situs, including right-sided stomach and spleen, multiple small splenenculi, reversed pancreas, midline liver, mispositioned transverse colon and abnormal anatomy of venous drainage. Follow-up BMA in 2004 (aged 28) showed MDS with normal karyotype (NK) with similar findings in 2015 (aged 39). Illumina Trusight myeloid panel showed an MDS-associated ZRSR2 variant (pGly438_Ser439insSerArg). Her daughter, IV-1, was born in 1996 at 28 weeks due to severe IUGR. Blood count at birth showed thrombocytopenia (90x109/l), presumed to be due to prematurity. However, aged two, she was noted to have anaemia (Hb 9.1g/dl) as well as thrombocytopenia (65x109/l). A BMA showed MDS with NK. She underwent an unrelated donor transplant at the age of four. Pre-transplant assessment noted that she had global developmental delay. She died two months post-transplant due to Epstein-Barr virus-related disease. The proband had five additional pregnancies. IV-3 was complicated by severe polyhydramnios and the pregnancy was terminated because of concerns regarding congenital anomalies. IV-5 is alive and well, aged 10. Informed consent was obtained from the UCL/UCLH Biobank for Studying Health and Disease. Whole exome sequencing was undertaken on stored bone marrow DNA of the proband from 2015, which identified two non-synonymous base changes in SAMD9 c.2958C> G ‒ resulting in S986C ‒ and c.1447C> G ‒ resulting in p.L483V (Fig 1B). In silico analyses of these variants suggested both mutations were likely to be pathogenic (Fig 1C). To determine if the mutations found in the proband were present in other family members, we undertook targeted deep sequencing of the regions surrounding the SAMD9 variants in IV-1 and the proband's mother (II-1) along with additional available samples from IV-5 and III-1 (Fig 1D). III-1, IV-1 and II-1 carried L483V at allele frequencies consistent with germline origin. The S986C variant was present only in samples from III-1. Here it was present at an allele frequency of 21% in the peripheral blood (2001) and 30·4% in the BMA from 2015. Matched buccal swab (2015) also showed this variant but at a lower allele frequency (17%). To determine whether this was due to blood contamination, we obtained nail clippings from III-1. This revealed the presence of the S986C mutation at a low frequency of 3·6%. This suggests that the proband underwent this reversion mutation in a germline mosaic manner early in development. A BMA from III-1 in 2015 was also analysed using SNP (single nucleotide polymorphism) array which showed no abnormalities. Specifically, we could not detect any evidence of UPD on chromosome 7 to account for the apparent reversion to 100% WT copies of chromosome 7 after 1998. However, as only 16% of cells in the 1998 sample carried this deletion, this would most likely be below the level of detection of the SNP array (20%). To identify if the two mutations found in SAMD9 in the proband were in cis or in trans, a PCR amplicon of SAMD9 encompassing both mutations (1701bp) was cloned and the colonies sequenced. Of the colonies, 33/65 (50·7%) carried L483V as expected, while 16/65 (24·6%) carried S986C (Fig 2B). Ninety-four per cent of colonies carrying S986C were WT for L483V, indicating that these mutations occurred in trans on opposite alleles. Two colonies carried both mutations, most likely representing crossover events. This is in contrast to previously described double mutations in SAMD9, where mutations have only been described in cis (Davidsson et al., 2018). In this report, we describe a four-generation pedigree of SAMD9 associated phenotypes, including AML, MDS and developmental anomalies. We highlight a novel germline SAMD9 mutation confirmed in three of the affected family members. The proband in this case also demonstrates a hitherto undescribed pattern of multiorgan involvement and congenital defects not previously associated with SAMD9 or SAMD9L mutations. The proband in this case displays genetic reversion from her inherited SAMD9 mutation by two different mechanisms; an in trans mutation of the other allele of SAMD9 that appears to have occurred during development in a germline mosaic manner, and may also explain the more extensive range of developmental anomalies observed in this case. In addition, the proband showed transient interstitial loss on chromosome 7q in 16% of cells early in the course of her MDS. Without the support of functional studies, it is tempting to speculate that the S986C mutation confers dominant negative activity against the putative gain-of-function L483V mutation and has clonal advantage over those with loss of 7q (Fig 2C). In summary, we describe a pedigree of SAMD9-mutated familial leukaemia predisposition with novel clinical and genetic features. CH wrote the manuscript and undertook the research, SG and YH undertook the research and edited the manuscript, RD and AR contributed to writing the manuscript, DCL, REG, MRM contributed to the research data and the manuscript, EP wrote the manuscript and conducted the study.

We compared the International Prognostic Index (IPI), Revised (R)-IPI and age-adjusted (aa)-IPI as prognostic indices for patients with diffuse large B-cell lymphoma (DLBCL) in the UK National Cancer Research Institute (NCRI) R-CHOP 14 versus 21 trial (N = 1080). The R-IPI and aa-IPI showed no marked improvement compared to the IPI for overall and progression-free survival, in terms of model fit or discrimination. Similar results were observed in exploratory analyses incorporating the Grupo Español de Linfomas/Transplante de Médula Ósea (GELTAMO)-IPI, where baseline β2-microglobulin data were available (N = 655). Although our findings support current use of the IPI, a novel prognostic tool to better delineate a high-risk DLBCL group in the rituximab era is needed.

Mary Gleeson, Nicholas Counsell, David Cunningham*, Anthony Lawrie, Laura Clifton-Hadley, Eliza Hawkes, Andrew McMillan, Kirit M Ardeshna, Cathy Burton, Nick Chadwick, Joanna Gambell, Paul Smith, Paul Mouncey, Christopher Pocock, John Radford, John Davies, Deborah Turner, Anton Kruger, Peter Johnson, David Linch The Royal Marsden Hospital, London and Surrey, United Kingdom, Cancer Research UK and UCL Cancer Trials Centre, UCL Cancer Institute, London, United Kingdom, Department of Oncology and Clinical Haematology, Austin Health, Heidelberg, Melbourne, Australia, Eastern Health, Melbourne Australia, Nottingham City Hospital, Nottingham, United Kingdom, University College London, London, United Kingdom, Mount Vernon Cancer Centre, Northwood, United Kingdom, HMDS, Leeds Teaching Hospitals NHS Trust, Leeds, United Kingdom, East Kent Hospitals, Canterbury, United Kingdom, The University of Manchester and the Christie NHS Foundation Trust, Manchester Academic Health Science Centre, Manchester, United Kingdom, Western General Hospital, Edinburgh, United Kingdom, Torbay Hospital, Torquay, United Kingdom, Royal Cornwall Hospital, Truro, United Kingdom, Cancer Research UK Centre, Southampton, United Kingdom

Abstract Loss-of-function mutations of EZH2, the enzymatic component of PRC2, have been associated with poor outcome and chemotherapy resistance in T-cell acute lymphoblastic leukemia (T-ALL). Using isogenic T-ALL cells, with and without CRISPR/Cas9–induced EZH2-inactivating mutations, we performed a cell-based synthetic lethal drug screen. EZH2-deficient cells exhibited increased sensitivity to structurally diverse inhibitors of CHK1, an interaction that could be validated genetically. Furthermore, small-molecule inhibition of CHK1 had efficacy in delaying tumor progression in isogenic EZH2-deficient, but not EZH2 wild-type, T-ALL cells in vivo, as well as in a primary cell model of PRC2-mutant ALL. Mechanistically, EZH2 deficiency resulted in a gene-expression signature of immature T-ALL cells, marked transcriptional upregulation of MYCN, increased replication stress, and enhanced dependency on CHK1 for cell survival. Finally, we demonstrate this phenotype is mediated through derepression of a distal PRC2-regulated MYCN enhancer. In conclusion, we highlight a novel and clinically exploitable pathway in high-risk EZH2-mutated T-ALL. Significance: Loss-of-function mutations of PRC2 genes are associated with chemotherapy resistance in T-ALL, yet no specific therapy for this aggressive subtype is currently clinically available. Our work demonstrates that loss of EZH2 activity leads to MYCN-driven replication stress, resulting in increased sensitivity to CHK1 inhibition, a finding with immediate clinical relevance. This article is highlighted in the In This Issue feature, p. 890