Computational flow cytometry provides accurate assessment of measurable residual disease in chronic lymphocytic leukaemia.

Undetectable measurable residual disease (MRD) is associated with favourable clinical outcomes in chronic lymphocytic leukaemia (CLL). While assessment is commonly performed using multiparameter flow cytometry (MFC), this approach is associated with limitations including user bias and expertise that may not be widely available. Implementation of unsupervised clustering algorithms in the laboratory can address these limitations and have not been previously reported in a systematic quantitative manner. We developed a computational pipeline to assess CLL MRD using FlowSOM. In the training step, a self-organising map was generated with nodes representing the full breadth of normal immature and mature B cells along with disease immunophenotypes. This map was used to detect MRD in multiple validation cohorts containing a total of 456 samples. This included an evaluation of atypical CLL cases and samples collected from two different laboratories. Computational MRD showed high correlation with expert analysis (Pearson's r > 0.99 for typical CLL). Binary classification of typical CLL samples as either MRD positive or negative demonstrated high concordance (>98%). Interestingly, computational MRD detected disease in a small number of atypical CLL cases in which MRD was not detected by expert analysis. These results demonstrate the feasibility and value of automated MFC analysis in a diagnostic laboratory.

Single-cell sequencing demonstrates complex resistance landscape in CLL and MCL treated with BTK and BCL2 inhibitors.

The genomic landscape of resistance to targeted agents (TAs) used as monotherapy in chronic lymphocytic leukemia (CLL) is complex and often heterogeneous at the patient level. To gain insight into the clonal architecture of acquired genomic resistance to Bruton tyrosine kinase (BTK) inhibitors and B-cell lymphoma 2 (BCL2) inhibitors in CLL, particularly in patients carrying multiple resistance mutations, we performed targeted single-cell DNA sequencing of 8 patients who developed progressive disease (PD) on TAs (either class). In all cases, analysis of single-cell architecture revealed mutual exclusivity between multiple resistance mutations to the same TA class, variable clonal co-occurrence of multiple mutations affecting different TAs in patients exposed to both classes, and a phenomenon of multiple independent emergences of identical nucleotide changes leading to canonical resistance mutations. We also report the first observation of established BCL2 resistance mutations in a patient with mantle cell lymphoma (MCL) following PD on sequential monotherapy, implicating BCL2 as a venetoclax resistance mechanism in MCL. Taken together, these data reveal the significant clonal complexity of CLL and MCL progression on TAs at the nucleotide level and confirm the presence of multiple, clonally independent, mechanisms of TA resistance within each individual disease context.

Characterisation of immune checkpoints in Richter syndrome identifies LAG3 as a potential therapeutic target.

Richter syndrome (RS), an aggressive lymphoma occurring in the context of chronic lymphocytic leukaemia/small lymphocytic lymphoma, is associated with poor prognosis when treated with conventional immunochemotherapy, therefore, improved treatments are required. Immune checkpoint blockade has shown efficacy in some B-cell malignancies and modest responses in early clinical trials for RS. We investigated the immune checkpoint profile of RS as a basis to inform rational therapeutic investigations in RS. Formalin-fixed, paraffin-embedded biopsies of RS (n = 19), de novo diffuse large B-cell lymphoma (DLBCL; n = 58), transformed indolent lymphomas (follicular [tFL], n = 16; marginal zone [tMZL], n = 24) and non-transformed small lymphocytic lymphoma (SLL; n = 15) underwent gene expression profiling using the NanoString Human Immunology panel. Copy number assessment was performed using next-generation sequencing. Immunohistochemistry (IHC) for LAG3 and PD-1 was performed. LAG3 gene expression was higher in RS compared to DLBCL (P = 0·0002, log2FC 1·96), tFL (P < 0·0001, log2FC 2·61), tMZL (P = 0·0004, log2FC 1·79) and SLL (P = 0·0057, log2FC 1·45). LAG3 gene expression correlated with the gene expression of human leukocyte antigen Class I and II, and related immune genes and immune checkpoints. IHC revealed LAG3 protein expression on both malignant RS cells and tumour-infiltrating lymphocytes. Our findings support the investigation of LAG3 inhibition to enhance anti-tumour responses in RS.

T cell receptor beta locus sequencing early post-allogeneic stem cell transplant identifies patients at risk of initial and recurrent cytomegalovirus infection.

Identification of patients at risk of initial & recurrent cytomegalovirus (CMV) reactivation following allogeneic stem cell transplant (alloSCT) may help guide prophylactic strategies. T-cell receptor beta (TRB) deep sequencing was used to identify and enumerate the T-cell repertoire harbouring TRB sequences with annotated specificity to CMV (pubCMVrep), as well as the overall T-cell receptor (TCR) repertoire diversity at day +30 & day +60 post-alloSCT for 65 patients. T-cells harbouring TRB sequences with annotated specificity for CMV were identifiable in all patients. 56% of patients required CMV treatment and 23% of the cohort developed recurrent CMV. PubCMVrep size at day +30 was not associated with reactivation, however amongst patients with antecedent CMV viremia a low day +60 pubCMVrep was associated with a greater incidence of recurrent CMV (75% vs. 21%, HR 6.16, 95% CI 1.29-29.40, P = 0.0008). Moreover, patients with high pubCMVrep only developed recurrent CMV in the setting of GVHD. Low TCR diversity at day +30 was associated with a greater incidence of initial CMV reactivation (71% vs. 22%, HR 5.39, 95% CI 1.70-17.09, p = 0.0002). pubCMVrep and TCR diversity are promising biomarkers to identify patients at risk of initial & recurrent CMV who may benefit from novel prophylactic strategies.

Novel genomic findings in multiple myeloma identified through routine diagnostic sequencing.

AIMS: Multiple myeloma is a genomically complex haematological malignancy with many genomic alterations recognised as important in diagnosis, prognosis and therapeutic decision making. Here, we provide a summary of genomic findings identified through routine diagnostic next-generation sequencing at our centre. METHODS: A cohort of 86 patients with multiple myeloma underwent diagnostic sequencing using a custom hybridisation-based panel targeting 104 genes. Sequence variants, genome-wide copy number changes and structural rearrangements were detected using an inhouse-developed bioinformatics pipeline. RESULTS: At least one mutation was found in 69 (80%) patients. Frequently mutated genes included TP53 (36%), KRAS (22.1%), NRAS (15.1%), FAM46C/DIS3 (8.1%) and TET2/FGFR3 (5.8%), including multiple mutations not previously described in myeloma. Importantly we observed TP53 mutations in the absence of a 17 p deletion in 8% of the cohort, highlighting the need for sequencing-based assessment in addition to cytogenetics to identify these high-risk patients. Multiple novel copy number changes and immunoglobulin heavy chain translocations are also discussed. CONCLUSIONS: Our results demonstrate that many clinically relevant genomic findings remain in multiple myeloma which have not yet been identified through large-scale sequencing efforts, and provide important mechanistic insights into plasma cell pathobiology.

Machine-Learning Algorithms Predict Graft Failure After Liver Transplantation.

BACKGROUND: The ability to predict graft failure or primary nonfunction at liver transplant decision time assists utilization of scarce resource of donor livers, while ensuring that patients who are urgently requiring a liver transplant are prioritized. An index that is derived to predict graft failure using donor and recipient factors, based on local data sets, will be more beneficial in the Australian context. METHODS: Liver transplant data from the Austin Hospital, Melbourne, Australia, from 2010 to 2013 has been included in the study. The top 15 donor, recipient, and transplant factors influencing the outcome of graft failure within 30 days were selected using a machine learning methodology. An algorithm predicting the outcome of interest was developed using those factors. RESULTS: Donor Risk Index predicts the outcome with an area under the receiver operating characteristic curve (AUC-ROC) value of 0.680 (95% confidence interval [CI], 0.669-0.690). The combination of the factors used in Donor Risk Index with the model for end-stage liver disease score yields an AUC-ROC of 0.764 (95% CI, 0.756-0.771), whereas survival outcomes after liver transplantation score obtains an AUC-ROC of 0.638 (95% CI, 0.632-0.645). The top 15 donor and recipient characteristics within random forests results in an AUC-ROC of 0.818 (95% CI, 0.812-0.824). CONCLUSIONS: Using donor, transplant, and recipient characteristics known at the decision time of a transplant, high accuracy in matching donors and recipients can be achieved, potentially providing assistance with clinical decision making.