Evidence-Based Learning

Journal Club &
Critical Appraisal

Structured summaries and critical appraisal of landmark haematology trials and high-impact publications — designed for trainees and practising clinicians.

8
Papers reviewed
4
BSH divisions covered
RCT
Level 1 evidence focus
This week · 25 May 2026
Effect snapshot — 4 new entries at a glance
Point estimates with 95% CI on a common ratio scale. Lower ratios favour the intervention; the dashed line is the null at 1.0.
null = 1.0 0.3 0.5 1.0 1.5 2.0 Tucidinostat + R-CHOP HR · DEL DLBCL HR 0.72 (0.54–0.96) Benzene exposure RR · lymphoid neoplasms RR 1.26 (1.18–1.35) Early ESAs · preterm RR · transfusion exposure RR 0.79 (0.76–0.83)
ISTH SSC UFH guidance — consensus document, not a ratio measure. Recommended anti-Xa target band 0.30–0.70 U/mL (different scale; see card below).
Filter:

Showing 10 entries

Malignant Haematology JAMA 2026

Tucidinostat plus R-CHOP versus R-CHOP in MYC/BCL2 Double-Expressor Diffuse Large B-Cell Lymphoma

Xu PP et al. JAMA 2026;335(19):1684–1693 — multicentre Phase III RCT, n = 423 (PMID: 42018318)

Key Findings

  • Event-free survival favoured tucidinostat: stratified HR 0.72 (95% CI 0.54–0.96; P = 0.02) — a 28% reduction in the composite of progression, relapse, death or new therapy
  • 2-year EFS 60.3% (tucidinostat + R-CHOP) vs 50.5% (R-CHOP alone)
  • Complete response rate 73.0% vs 61.8% (absolute difference 11.1%, 95% CI 2.3–20.0%)
  • Toxicity higher in the tucidinostat arm but described as manageable with supportive care
Clinical Takeaway

The first positive Phase III trial of an epigenetic modifier added to R-CHOP in MYC/BCL2 double-expressor DLBCL — a high-risk subgroup with historically poor outcomes on R-CHOP alone. Tucidinostat is not licensed by the MHRA and not recommended by NICE for first-line DLBCL; UK adoption would need NICE technology appraisal. Pending BSH review, this remains the strongest positive epigenetic signal in DEL DLBCL to date and is worth raising at MDT for high-risk subgroups.

Effect — Event-free survival (lower HR favours tucidinostat)
HR 1.0 (null) 0.3 0.5 1.0 1.5 2.0 HR 0.72 (0.54–0.96)
Critical Appraisal Note▾

Multicentre double-blind RCT with 41.3-month median follow-up. Three caveats: enrolment was exclusively at 40 Chinese centres (limits external validity to UK populations); DEL was defined by IHC co-expression, not FISH-confirmed double-hit rearrangement — these are overlapping but distinct populations; overall survival not yet mature.

1B Level 1 Evidence — Phase III RCT
Malignant Haematology Occup Environ Med 2026

Benzene Exposure and Lymphoid Neoplasm Risk under WHO-HAEM5 Reclassification

Park HW et al. Occup Environ Med 2026;83(2):113–120 — Systematic review and meta-analysis, 95 studies (PMID: 42082363)

Key Findings

  • Overall lymphoid neoplasm risk with benzene exposure: RR 1.26 (95% CI 1.18–1.35)
  • B-cell neoplasm risk: RR 1.26 (95% CI 1.16–1.37) — including mature B-cell, Hodgkin lymphoma and plasma-cell neoplasms
  • T-cell / NK-cell neoplasm risk not elevated overall
  • Occupational exposures carried higher risks with less heterogeneity than environmental exposures
Clinical Takeaway

Take an occupational history at every new lymphoma diagnosis. Where benzene-rich industries (petrochemical, printing, paint, rubber, fuel tanker) are involved, document carefully — for clinical risk profiling and for any medicolegal or industrial-injury pathway. Confirms that the 2022 WHO-HAEM5 framework adds useful resolution to historical occupational-cancer epidemiology.

Effect — Lymphoid neoplasm risk (RR > 1 indicates higher risk)
RR 1.0 (null) 0.5 1.0 1.5 2.0 RR 1.26 (1.18–1.35)
Critical Appraisal Note▾

Heterogeneous exposure measurement across the 95 primary studies; residual confounding from co-exposures (other solvents, smoking). Modest effect sizes susceptible to publication bias. Reclassification under WHO-HAEM5 was retrospective and depended on the granularity of the original diagnoses.

3A Level 3 Evidence — SR of observational studies
Haemostasis & Thrombosis J Thromb Haemost 2026

ISTH SSC Guidance on Monitoring and Managing Unfractionated Heparin (Part 1)

Gouin-Thibault I et al. J Thromb Haemost 2026 — ISTH SSC Delphi consensus guidance (PMID: 42162738)

Key Findings

  • Suggested anti-Xa target 0.30–0.70 U/mL for most indications (VTE, arterial events, mechanical heart valves)
  • Anti-Xa preferred over aPTT — fewer interferences, no need for a locally-derived therapeutic range
  • Assays without dextran sulphate or added antithrombin are preferred, particularly to avoid overestimating UFH activity after protamine reversal
  • 3.2% citrate tubes recommended; centrifugation within 1 hour and testing within 10 minutes ideal (1 hour acceptable). Antithrombin supplementation not routinely advised for acquired deficiency
Clinical Takeaway

The first multinational consensus framework for standardising UFH monitoring. Many UK trusts still rely on aPTT — this guidance gives labs a clear direction of travel toward anti-Xa-based monitoring with a defined international target range. UK haematology and ICU leads should benchmark local protocols against this consensus and document any deviation. Pending BSH endorsement.

Recommended anti-Xa target band (U/mL)
0.0 0.20 0.40 0.60 0.80 1.00 U/mL 0.30 – 0.70 U/mL (target band)
Critical Appraisal Note▾

Consensus-driven rather than trial-driven; outcome data for several recommendations remain sparse. Weight-based dosing nomograms are recommended but no single nomogram is preferred. Special populations and periprocedural use are covered in a forthcoming Part 2.

5 Expert consensus — ISTH SSC Delphi process
Transfusion Medicine Cochrane 2026

Early Erythropoiesis-Stimulating Agents in Preterm or Low-Birthweight Infants — Updated Cochrane Review

Anarna K, Fiander M, Mitra S. Cochrane Database Syst Rev 2026;5(5):CD004863 — 37 RCTs, n = 6,724 (PMID: 42170845) — cross-tag: General Haematology

Key Findings

  • Red-cell transfusion exposure reduced — RR 0.79 (95% CI 0.76–0.83; low certainty) — primary reason for the Transfusion cross-tag
  • Necrotising enterocolitis reduced — RR 0.74 (0.62–0.87; moderate certainty)
  • Severe intraventricular haemorrhage reduced — RR 0.70 (0.56–0.89; moderate certainty)
  • Mortality during initial hospital stay — probably no difference, RR 0.93 (0.80–1.09; moderate certainty)
  • Retinopathy of prematurity (any stage) — no difference (high certainty)
Clinical Takeaway

Consistent with existing UK neonatal practice — supports continued use of early ESAs for transfusion reduction and NEC / severe IVH prevention in preterm and low-birthweight infants. Mortality is not changed. From a transfusion-medicine perspective, a 21% relative reduction in red-cell transfusion exposure across a 6,724-infant evidence base is a meaningful blood-conservation signal worth quoting in transfusion governance and neonatal committee discussions.

Effect — four key outcomes (lower RR favours early ESAs)
RR 1.0 (null) 0.5 0.75 1.0 1.25 1.5 Transfusion exposure RR 0.79 (0.76–0.83) NEC RR 0.74 (0.62–0.87) Severe IVH RR 0.70 (0.56–0.89) Mortality RR 0.93 (0.80–1.09)
Critical Appraisal Note▾

Variable heterogeneity across outcomes; certainty graded down for inconsistency and risk of bias on several endpoints. Subgroup data for darbepoetin is sparser than for erythropoietin. The neurodevelopmental signal is low-certainty and should be quoted with that caveat. The transfusion-exposure outcome is low-certainty despite the precise effect estimate.

1A Level 1 Evidence — Cochrane systematic review
Malignant Haematology NEJM 2022

Polatuzumab Vedotin in Previously Untreated Diffuse Large B-Cell Lymphoma

Tilly H et al. N Engl J Med 2022;386:351–363 — POLARIX Trial (Phase III RCT)

Key Findings

  • 2-year progression-free survival: 76.7% with Pola-R-CHP vs 70.2% with R-CHOP (statistically significant)
  • No significant overall survival difference at primary analysis — longer follow-up awaited
  • Peripheral neuropathy rates were comparable between arms (52.9% vs 53.9% any grade), with fewer dose reductions required in the polatuzumab arm (4.4% vs 8.0%)
Clinical Takeaway

Pola-R-CHP is now an established front-line option for intermediate- and high-risk DLBCL, replacing vincristine with a targeted antibody-drug conjugate. The absence of an OS benefit at primary analysis means R-CHOP remains reasonable, particularly in lower-risk or elderly patients.

Critical Appraisal Note

Double-blind RCT with robust design. Key limitation: the OS benefit was not demonstrated at primary analysis. Patient selection (IPI ≥2) limits generalisability to low-risk disease. Five-year follow-up data presented at ASH 2024 showed sustained PFS advantage.

1A Level 1 Evidence — Phase III RCT
Haemostasis & Thrombosis NEJM 2020

Apixaban for the Treatment of Venous Thromboembolism Associated with Cancer

Agnelli G et al. N Engl J Med 2020;382:1599–1607 — CARAVAGGIO Trial (Phase III RCT)

Key Findings

  • Recurrent VTE: 5.6% with apixaban vs 7.9% with dalteparin (HR 0.63; 95% CI 0.37–1.07; P<0.001 for non-inferiority)
  • Major bleeding: 3.8% vs 4.0% — no significant increase with apixaban (HR 0.82; 95% CI 0.40–1.69; P=0.60)
  • Oral administration of apixaban was non-inferior to subcutaneous dalteparin across all cancer subtypes analysed
Clinical Takeaway

Apixaban is non-inferior to dalteparin for cancer-associated VTE without a significantly increased bleeding risk, making it a practical oral alternative to low-molecular-weight heparin in eligible patients. Caution remains warranted in GI and genitourinary malignancies due to historically higher mucosal bleeding rates with DOACs in these tumour types.

Critical Appraisal Note

Well-designed investigator-initiated RCT. Non-inferiority design: the CI upper bound (1.07) does not exclude a small increase in recurrent VTE. Patients with GI and GU cancers at higher bleeding risk were not excluded — clinicians should apply individual risk stratification in these groups.

1A Level 1 Evidence — Phase III RCT
Malignant Haematology NEJM 2018

Venetoclax–Rituximab in Relapsed or Refractory Chronic Lymphocytic Leukaemia

Seymour JF et al. N Engl J Med 2018;378:1107–1120 — MURANO Trial (Phase III RCT)

Key Findings

  • At median 36-month follow-up: PFS markedly superior with VenR vs bendamustine-rituximab (HR 0.16; 95% CI 0.12–0.23)
  • Overall survival also superior (HR 0.50; 95% CI 0.30–0.85) — a rare demonstration of OS benefit in r/r CLL
  • Undetectable MRD at end of combination therapy: 62% with VenR vs 13% with BR — establishing fixed-duration treatment as a feasible strategy
Clinical Takeaway

MURANO established fixed-duration venetoclax-rituximab as a preferred option for relapsed/refractory CLL, particularly for patients unsuitable for indefinite BTK inhibitor therapy. The high rate of undetectable MRD supports a time-limited treatment strategy with potential for treatment-free remission.

Critical Appraisal Note

Phase III RCT with meaningful OS benefit — a high bar in the CLL field. Comparator arm (BR) may be considered suboptimal by current standards. Tumour lysis syndrome risk requires a structured ramp-up protocol, limiting use to centres with appropriate monitoring infrastructure.

1A Level 1 Evidence — Phase III RCT
General Haematology Coming Soon

Iron Deficiency — Intravenous vs Oral Iron in Anaemia Management

Critical appraisal of RCT evidence comparing IV and oral iron formulations in iron deficiency anaemia — covering efficacy, adverse effects, and practical prescribing guidance.

Transfusion Medicine Coming Soon

Restrictive vs Liberal Red Cell Transfusion Thresholds — Evidence Review

Summary and appraisal of landmark trials including TRICC and TRISS examining haemoglobin transfusion triggers across critical care, cardiac, and haematology patient populations.

Haemostasis & Thrombosis Evidence review 2026

Extended Anticoagulation in Unprovoked VTE — Balancing Recurrence vs Bleeding Risk

A structured review of AMPLIFY-EXT (apixaban) and EINSTEIN CHOICE (rivaroxaban), with the three risk-stratification tools (HERDOO2, DASH, Vienna) and interactive calculators.

Key Findings

  • Apixaban 2.5 mg BD (AMPLIFY-EXT) reduced 12-month recurrence from 8.8% to 1.7% versus placebo with no increase in major bleeding
  • Rivaroxaban 10 mg OD (EINSTEIN CHOICE) reduced recurrence about 3-fold versus aspirin 100 mg (HR 0.26, 95% CI 0.14–0.47); aspirin is no longer the default fallback
  • HERDOO2 is the only tool with prospective management validation; women scoring 0–1 had about 3.0% per patient-year recurrence after stopping (Rodger 2017, BMJ)
1A Level 1 Evidence — Phase III RCTs + validated prediction tools