NEJM This Week — May 28, 2026

This week's issue features three randomized controlled trials — HI-PEITHO (catheter-directed fibrinolysis in intermediate-risk PE), B-Well 1 & 2 (bepirovirsen functional cure in chronic HBV), and HERIZON-GEA-01 (zanidatamab in HER2+ gastroesophageal cancer) — plus a high-yield Leishmaniasis review, a teaching case of tumefactive demyelination in an older adult, and non-randomized studies including the BioVAT-HF engineered heart muscle trial and a pediatric Chiari I surgery cluster-RCT.

Pulmonary / Critical CareHI-PEITHO · RCTHigh yield · IM/ID

Ultrasound-Facilitated Catheter-Directed Fibrinolysis for Intermediate-Risk Pulmonary Embolism

Population: 544 adults (age 18–80) with acute, intermediate-risk PE confirmed by CT pulmonary angiography — RV:LV end-diastolic diameter ratio ≥1.0 + elevated troponin + ≥2 of: SBP ≤110 mmHg, HR ≥100 bpm, RR >20, or hypoxemia. Excluded: persistent hemodynamic instability (high-risk PE), age >80. Mean age 58.2 yr; 42.6% women.
Intervention: Ultrasound-facilitated, catheter-directed fibrinolysis (EkoSonic system) with alteplase (mean 8.85 mg unilateral / 16.92 mg bilateral) plus anticoagulation; initiated within 2 hr of randomization.
Comparison: Anticoagulation alone (current standard of care).
Outcome: Primary composite (PE-related death + cardiorespiratory decompensation/collapse + symptomatic PE recurrence) at 7 days: 4.0% vs 10.3% (RR 0.39; 95% CI 0.20–0.77; P=0.005). Driven almost entirely by lower cardiorespiratory decompensation (3.7% vs 10.3%). Major bleeding at 30 days: 4.1% vs 3.0% (P=0.64); no intracranial hemorrhage in either group.

Clinical takeaway. In intermediate-risk PE with cardiorespiratory distress criteria, catheter-directed fibrinolysis (CDT) with ultrasound facilitation reduced the composite of 7-day hemodynamic collapse/PE death/recurrence by 61% relative (NNT ≈ 16) with no significant increase in major bleeding or intracranial hemorrhage — a meaningful improvement over the systemic fibrinolysis data from PEITHO. The editorial (Spyropoulos and Vedantham) endorses a lower threshold for CDT in intermediate-high risk PE but urges caution: NEWS ≥9 (used to define decompensation) is not widely used in North America; a large number of screened patients were excluded for unclear reasons; and the trial was industry-funded (Boston Scientific). Short-term 6-min walk distances and RV/LV ratio changes were unimpressive; 12-month follow-up data are pending.

Infectious DiseaseB-Well 1 & 2 · RCTHigh yield · IM/ID

Bepirovirsen for Functional Cure in Chronic Hepatitis B (B-Well 1 & 2)

Population: 1,838 adults with noncirrhotic chronic HBV infection on stable nucleoside/nucleotide analogue (NA) therapy ≥6 months; HBsAg 100–3000 IU/mL; HBV DNA <90 IU/mL; ALT ≤2× ULN. Excluded: cirrhosis, HCV/HIV/HDV coinfection, history of vasculitis/autoimmune disease, platelets <140,000/mm³. Randomized 2:1 in each trial (B-Well 1: 650/328; B-Well 2: 570/286).
Intervention: Bepirovirsen 300 mg subcutaneous weekly × 24 weeks (plus continuing background NA); eligible patients discontinued NA at week 48 if HBV DNA undetectable and HBsAg undetected at weeks 24–46.
Comparison: Placebo subcutaneous × 24 weeks plus NA; no NA discontinuation.
Outcome: Functional cure (HBV DNA <LLOQ + HBsAg undetected for ≥24 wks after stopping all HBV therapy) at week 72: 20% (B-Well 1) and 19% (B-Well 2) vs 0% in both placebo arms (P<0.001 for both). In lower HBsAg stratum (≤1000 IU/mL): 25–28% functional cure rates. Grade ≥3 adverse events: 16% vs 3%; ALT elevation 6%; injection-site reactions 53% vs 14% (mostly mild); transient platelet and eGFR decreases resolved by week 72.

Clinical takeaway. Bepirovirsen, an antisense oligonucleotide targeting all HBV transcripts, achieved functional cure (HBsAg loss + undetectable DNA off all treatment) in ~20% of NA-suppressed chronic HBV patients — compared to essentially 0% with continued NA therapy. This represents a landmark advance; no prior single investigational agent has achieved this functional cure rate in phase 3 trials. The editorial (Anna Lok) frames this as "a major step" but notes important limits: the 20% cure rate still leaves 80% without cure; the benefit is substantially lower in patients with HBsAg >1000 IU/mL (only 5–10%); cirrhosis, HIV/HCV coinfection, and HBsAg >3000 IU/mL were excluded. ALT flares, platelet drops, and eGFR declines require stringent monitoring (every 1–2 weeks during treatment). Durability beyond week 72 is being assessed in the B-Sure trial.

OncologyHERIZON-GEA-01 · RCTLower yield · IM/ID

Zanidatamab ± Tislelizumab vs. Trastuzumab in HER2-Positive Gastroesophageal Adenocarcinoma

Population: 914 adults with previously untreated, centrally confirmed HER2-positive (IHC 3+ or IHC 2+/ISH+) advanced gastroesophageal adenocarcinoma (gastric, GEJ, or esophageal); ECOG 0–1. No prior systemic treatment for advanced disease. Randomized 1:1:1 across three arms.
Intervention: Arm A: Zanidatamab (bispecific anti-HER2) + tislelizumab (anti–PD-1) + chemotherapy (CAPOX or FOLFCIS). Arm B: Zanidatamab + chemotherapy.
Comparison: Trastuzumab + chemotherapy (standard of care).
Outcome: Median PFS: zanidatamab–tislelizumab–chemo 12.4 months vs trastuzumab–chemo 8.1 months (HR 0.63; P<0.001). Zanidatamab–chemo alone also 12.4 months vs 8.1 months (HR 0.65; P<0.001). OS (interim): zanidatamab–tislelizumab–chemo 26.4 months vs 19.2 months (HR 0.72; P=0.004). OS for zanidatamab–chemo alone (24.4 months) did not reach significance (P=0.06) at this interim. Grade ≥3 adverse events: 83.3% (triple) / 73.8% (doublet) vs 74.5% (control).

Clinical takeaway. In previously untreated HER2+ advanced gastroesophageal adenocarcinoma, both zanidatamab-containing arms significantly improved progression-free survival versus trastuzumab + chemotherapy, with the triple combination also demonstrating a significant OS benefit at interim analysis. This is subspecialty GI oncology territory; lower-yield for the generalist IM/ID clinician but informative for GI/oncology fellows regarding the emerging role of bispecific HER2 antibodies in this setting. Diarrhea was the most common high-grade event (24.8% / 20% vs 12.9% with control).

Infectious DiseaseReview · High yield · ID fellow

Leishmaniasis

A high-yield tropical/ID review with direct clinical relevance for ID fellows and military medicine. Key points:

•Visceral leishmaniasis (VL; kala-azar) has 75–95% mortality if untreated, reduced to 3–10% with treatment; caused by L. donovani / L. infantum (transmitted by sand flies). •VL is an opportunistic infection in HIV (CD4 <200 → severe atypical disease). WHO recommends liposomal amphotericin B for first episode, then secondary prophylaxis until CD4 >200–350 and viral load controlled. •Recent advance: Combination therapy (paromomycin + miltefosine) for East African VL — more cost-effective and lower toxicity than paromomycin + antimonials. •Diagnosis has shifted to molecular methods (PCR) on tissue (skin biopsy, bone marrow). The rK39 serologic test (Kalazar Detect) is FDA-approved but has reduced sensitivity in East Africa.

Teaching Case · Case Records of the MGH

Case 15-2026: A 64-Year-Old Woman with Fatigue, Memory Impairment, and Falls · Neurology

High yield · IM/ID · subacute neuro decline

Key learning point. Final diagnosis: tumefactive demyelination / late-onset multiple sclerosis. Expansile white-matter lesions with open-ring enhancement and a striking clinical–radiologic mismatch (many lesions, only modest deficits), no restricted diffusion, and a later central vein sign confirmed demyelination on biopsy (myelin loss with axonal preservation; 8 CSF-restricted oligoclonal bands). Open-ring enhancement is >95% specific for demyelination versus tumor or abscess. Approximately 10% of new MS diagnoses are now in patients >50. Treated with IV methylprednisolone, then anti-CD20 (ublituximab).

Also in this issue (not randomized, so not summarized as trials): BioVAT-HF (Cardiology) — Open-label phase 1–2 (n=20) of epicardially transplanted engineered heart muscle (cardiomyocytes + stromal cells from induced pluripotent stem cells) in advanced HFrEF (LVEF ≤35%) refractory to GDMT; interim signal of wall-thickness increase and modest LVEF improvement, no control group. Chiari I / Syringomyelia Surgery (Pediatric Neurosurgery) — Cluster-randomized trial (n=162, age ≤21) of posterior fossa decompression with duraplasty (PFD-D) vs. PFD alone; greater syrinx reduction with PFD-D but more surgical complications (14% vs 6%, not significant) and more repeat decompression with PFD alone (14% vs 3%).

Summaries are original paraphrases prepared for educational use; figures are drawn from the published full text. Read the full articles at NEJM.org (subscription required). Trial registrations: HI-PEITHO (NCT04790370), B-Well 1 (NCT05630807), B-Well 2 (NCT05630820). Citation: N Engl J Med 2026; Vol. 394, No. 20. DOIs: 10.1056/NEJMoa2516567, 10.1056/NEJMoa2515131, 10.1056/NEJMra2403309, 10.1056/NEJMcpc2517865, 10.1056/NEJMoa2513525, 10.1056/NEJMoa2402821.