David T. Yeung

Haematopoietic stem cell transplantation (HSCT) is a curative approach for blood cancers, yet its efficacy is undermined by a range of acute and chronic complications. In light of mounting evidence to suggest that these complications are linked to a dysbiotic gut microbiome, we aimed to evaluate the feasibility of faecal microbiota transplantation (FMT) delivered during the acute phase after HSCT. Of note, this trial opted for FMT prepared using the individual’s own stool (autologous FMT) to mitigate the risks of disease transmission from a donor stool. Adults (>18 years) with multiple myeloma were recruited from a single centre. The stool was collected prior to starting first line therapy. Patients who progressed to HSCT were offered FMT via 3 × retention enemas before day +5 (HSCT = day 0). The feasibility was determined by the recruitment rate, number and volume of enemas administered, and the retention time. Longitudinally collected stool samples were also collected to explore the influence of auto-FMT using 16S rRNA gene sequencing. n = 4 (2F:2M) participants received auto-FMT in 12 months. Participants received an average of 2.25 (1–3) enemas 43.67 (25–50) mL total, retained for an average of 60.78 (10–145) min. No adverse events (AEs) attributed to the FMT were identified. Although the minimum requirements were met for the volume and retention of auto-FMT, the recruitment was significantly impacted by the logistical challenges of the pretherapy stool collection. This ultimately undermined the feasibility of this trial and suggests that third party (donor) FMT should be prioritised.

Haematopoietic stem cell transplantation (HSCT) is a curative approach for blood cancers, yet its efficacy is undermined by a range of acute and chronic complications. In light of mounting evidence to suggest that these complications are linked to a dysbiotic gut microbiome, we aimed to evaluate the feasibility of faecal microbiota transplantation (FMT) delivered during the acute phase after HSCT. Of note, this trial opted for FMT prepared using the individual’s own stool (autologous FMT) to mitigate risks of disease transmission from donor stool. Adults (>18 years) with multiple myeloma were recruited from a single centre. Stool was collected prior to starting first-line therapy. Patients that progressed to HSCT were offered FMT via 3 x re-tention enemas before day +5 (HSCT = day 0). Feasibility was determined by recruitment rate, number and volume of enemas administered, and retention time. Longitudinally collected stool samples were also col-lected to explore the influence of auto-FMT using 16S rRNA gene sequencing. N=4 (2F:2M) participants received auto-FMT in 12 months. Participants received an average of 2.25(1-3) enemas (43.67(25-50)mL total, retained for an average of 60.78(10-145)minutes). No AEs, attributed to the FMT, were identified. Although minimum requirements were met for the volume and retention of auto-FMT, recruitment was significantly impacted by the logistical challenges of pre-therapy stool collection. This ultimately under-mined the feasibility of this trial and suggests that third party (donor) FMT should be prioritised.

B-cell acute lymphoblastic leukaemia (B-ALL) is characterised by diverse genomic alterations, the most frequent being gene fusions detected via transcriptomic analysis (mRNA-seq). Due to its hypervariable nature, gene fusions involving the Immunoglobulin Heavy Chain (IGH) locus can be difficult to detect with standard gene fusion calling algorithms and significant computational resources and analysis times are required. We aimed to optimize a gene fusion calling workflow to achieve best-case sensitivity for IGH gene fusion detection. Using Nextflow, we developed a simplified workflow containing the algorithms FusionCatcher, Arriba, and STAR-Fusion. We analysed samples from 35 patients harbouring IGH fusions (IGH::CRLF2 n = 17, IGH::DUX4 n = 15, IGH::EPOR n = 3) and assessed the detection rates for each caller, before optimizing the parameters to enhance sensitivity for IGH fusions. Initial results showed that FusionCatcher and Arriba outperformed STAR-Fusion (85–89% vs. 29% of IGH fusions reported). We found that extensive filtering in STAR-Fusion hindered IGH reporting. By adjusting specific filtering steps (e.g., read support, fusion fragments per million total reads), we achieved a 94% reporting rate for IGH fusions with STAR-Fusion. This analysis highlights the importance of filtering optimization for IGH gene fusion events, offering alternative workflows for difficult-to-detect high-risk B-ALL subtypes.

RNA sequencing provides a snapshot of the functional consequences of genomic lesions that drive acute lymphoblastic leukemia (ALL). The aims of this study were to elucidate diagnostic associations (via machine learning) between mRNA-seq profiles, independently verify ALL lesions and develop easy-to-interpret transcriptome-wide biomarkers for ALL subtyping in the clinical setting. A training dataset of 1279 ALL patients from six North American cohorts was used for developing machine learning models. Results were validated in 767 patients from Australia with a quality control dataset across 31 tissues from 1160 non-ALL donors. A novel batch correction method was introduced and applied to adjust for cohort differences. Out of 18,503 genes with usable expression, 11,830 (64%) were confounded by cohort effects and excluded. Six ALL subtypes (ETV6::RUNX1, KMT2A, DUX4, PAX5 P80R, TCF3::PBX1, ZNF384) that covered 32% of patients were robustly detected by mRNA-seq (positive predictive value ≥ 87%). Five other frequent subtypes (CRLF2, hypodiploid, hyperdiploid, PAX5 alterations and Ph-positive) were distinguishable in 40% of patients at lower accuracy (52% ≤ positive predictive value ≤ 73%). Based on these findings, we introduce the Allspice R package to predict ALL subtypes and driver genes from unadjusted mRNA-seq read counts as encountered in real-world settings. Two examples of Allspice applied to previously unseen ALL patient samples with atypical lesions are included.

DUX4-rearrangement (DUX4r) is a recently discovered recurrent genomic lesion reported in 4–7% of childhood B cell acute lymphoblastic leukaemia (B-ALL) cases. This subtype has favourable outcomes, especially in children and adolescents treated with intensive chemotherapy. The fusion most commonly links the hypervariable IGH gene to DUX4 a gene located within the D4Z4 macrosatellite repeat on chromosome 4, with a homologous polymorphic repeat on chromosome 10. DUX4r is cryptic to most standard diagnostic techniques, and difficult to identify even with next generation sequencing assays. This review summarises the clinical features and molecular genetics of DUX4r B-ALL and proposes prospective new diagnostic methods.

An 80-year-old man has newly diagnosed chronic myeloid leukemia. His BM and blood examination at diagnosis confirms chronic-phase disease, with the Philadelphia chromosome as the sole cytogenetic abnormality. He has intermediate Sokal and Hasford risk,¹  and is started on imatinib 600 mg once daily. He lives 5 hours away from the nearest specialist hematology service and prefers followup with his local physician, who cannot perform BM examinations. In patients such as this, is it acceptable to monitor his therapeutic response solely with molecular studies of his peripheral blood?

An 80-year-old man has newly diagnosed chronic myeloid leukemia. His BM and blood examination at diagnosis confirms chronic-phase disease, with the Philadelphia chromosome as the sole cytogenetic abnormality. He has intermediate Sokal and Hasford risk,(1) and is started on imatinib 600 mg once daily. He lives 5 hours away from the nearest specialist hematology service and prefers followup with his local physician, who cannot perform BM examinations. In patients such as this, is it acceptable to monitor his therapeutic response solely with molecular studies of his peripheral blood?David T. Yeung and Susan Branfor