Five papers from July 14-21: OPTIMA-AF leads the PubMed window

Five high-impact papers entered PubMed's July 14-21 EDAT window: two randomized trials, a pooled prognostic cohort, a phase 1 lymphoma combination study, and a first-in-human ALS program. OpenAlex reports 0 cited-by works for each paper at the July 21 cutoff, so early citation momentum does not separate the finalists. The order below uses journal impact tier, design strength, sample size, and direct clinical or research utility as tie-breakers. The ranks are therefore an early evidence-weighted triage order, not a mature citation leaderboard. 1 2 3 4 5

Paper link: PubMed record · The Lancet DOI

What the paper did. OPTIMA-AF randomized adults with non-valvular atrial fibrillation undergoing intravascular-imaging-guided PCI at 75 Japanese sites to one month of a DOAC plus a P2Y12 inhibitor followed by DOAC monotherapy, or 12 months of dual therapy followed by DOAC monotherapy. Investigators and participants were unmasked; an independent committee adjudicated clinical events while masked to treatment. The PubMed record was dated July 16, 2026. 6

Primary result. Death or thromboembolic events occurred in 29 of 542 patients in the one-month group versus 23 of 537 in the 12-month group, a Kaplan-Meier estimate of 5.4% versus 4.3% and HR 1.25 (95% CI 0.73-2.17). The one-month strategy was non-inferior for efficacy. Major or clinically relevant non-major bleeding occurred in 24 versus 47 patients, 4.5% versus 8.8%, HR 0.50 (95% CI 0.30-0.81; P=.0041 for superiority). 6

Evidence strength and limitation. This is the most practice-facing study in the set: randomized, event-adjudicated, and large enough to test a concrete antithrombotic duration question. The event rate was lower than expected, and the fixed absolute non-inferiority margin makes the efficacy result worth reading in full. The population was predominantly older Japanese adults with chronic coronary syndrome, so extrapolation to acute coronary syndromes or higher-thrombotic-risk groups is limited.

Clinical implication. For the trial population, shortening dual therapy to one month reduced bleeding without a demonstrated loss of protection from death or thromboembolism. It supports a shorter default strategy where ischemic risk is acceptable, but it does not eliminate the need to individualize treatment after PCI.

Author affiliation and funding. The retrieved PubMed and Crossref metadata did not expose a lead-author affiliation. Funding was reported as Abbott Medical Japan. 6

Paper link: PubMed record · The Lancet DOI

What the paper did. This 2 x 2 factorial, open-label randomized trial enrolled infants aged 28-365 days with severe HIV-associated pneumonia across 19 hospitals in Cote d'Ivoire, Malawi, Mozambique, Uganda, Zambia, and Zimbabwe. The reported comparison was standard care with or without 15 days of oral valganciclovir at 16 mg/kg every 12 hours; tuberculosis treatment was the second factorial intervention. The PubMed record was dated July 18, 2026. 7

Primary result. At day 15, mortality was 64/276 (23%) with valganciclovir versus 76/282 (27%) without, rate ratio 0.81 (95% CI 0.61-1.08; P=.15). At 12 months, mortality was 119/276 (43%) versus 134/282 (48%), estimate 0.88 (95% CI 0.74-1.05; P=.15). A time-varying effects model found an adjusted HR of 0.60 (95% CI 0.41-0.87; P=.0063) at day 15, but the one-year HR was 0.79 (95% CI 0.62-1.01; P=.068). Severe adverse events were not more common, OR 0.64 (95% CI 0.35-1.16). 7

Evidence strength and limitation. The geographic reach and randomized design are major strengths. The main result is not statistically significant at either prespecified mortality time point, while the time-varying model suggests an early hazard difference that attenuates over follow-up. Loss to follow-up was 9% in valganciclovir groups and 5% in groups without it. That combination makes this a signal-generating result rather than a stand-alone mandate for empiric treatment.

Clinical implication. The study gives clinicians a reason to examine the timing of mortality and the biological plausibility of empiric CMV treatment in this population, but the neutral primary comparisons leave practice change dependent on replication and context. The result is most relevant where severe HIV-associated pneumonia remains a high-mortality presentation and virologic or radiographic confirmation is delayed.

Author affiliation and funding. Lead author Cinta Moraleda is affiliated with Instituto de Investigación Sanitaria Hospital 12 de Octubre and Hospital Universitario 12 de Octubre in Madrid; the author group spans the participating African hospitals and research centers. Funding was reported as the European and Developing Countries Clinical Trials Partnership. 7

Paper link: PubMed record · JAMA DOI

What the paper did. This longitudinal cohort pooled harmonized data from six observational and clinical-trial cohorts in North America, Japan, and Australia. It included 2,684 cognitively unimpaired older adults, followed for a median of 5.4 years; 478 participants progressed to cognitive impairment. The PubMed record was dated July 15, 2026. 8

Primary result. Each 1-SD increase in baseline plasma p-tau217 was associated with a higher risk of progression, HR 1.38 (95% CI 1.30-1.46), and HR 1.32 (95% CI 1.24-1.41) after adjustment including amyloid PET Centiloids. Five-year absolute risk was 24% (95% CI 20%-28%) for the high p-tau217 group and 38% (95% CI 33%-43%) for the very-high group. The very-high group declined by -0.07 latent PACC units per year (95% CI -0.10 to -0.05), compared with 0.03 units per year (95% CI 0.02-0.04) in the low group. 8

Evidence strength and limitation. The paper improves on a simple association by giving time-specific absolute risk estimates and by testing persistence after amyloid PET adjustment. The cohorts were selected observational and trial populations, not an unselected primary-care sample. The authors therefore call for external validation before individual prognostic use, a limitation that matters as blood biomarkers move toward routine testing.

Clinical implication. The most immediate use is trial enrichment and prognostic-model development. A p-tau217 value can stratify risk in this research setting, but these estimates should not be read as a stand-alone diagnosis or as a clinical prediction rule for cognitively unimpaired people.

Author affiliation and funding. Lead author Rachel F. Buckley is affiliated with the Department of Neurology at Mass General Brigham and the Melbourne School of Psychological Sciences at the University of Melbourne. Retrieved metadata lists NIH/NIA and Alzheimer's Association support across the contributing cohort programs; no single consolidated funding statement was exposed in the abstract view. 8

Paper link: PubMed record · Nature Medicine full article

What the paper did. Part 2 of this open-label, nonrandomized phase 1 study escalated the CD19-4-1BBL co-stimulatory molecule englumafusp alfa with glofitamab in relapsed or refractory B-cell non-Hodgkin lymphoma. Obinutuzumab pretreatment and glofitamab step-up dosing preceded up to 11 cycles of combination treatment. The study enrolled 134 patients: 109 with aggressive and 25 with indolent B-cell NHL. The PubMed record was dated July 17, 2026. 9

Primary result. The maximum tolerated dose was not reached; one dose-limiting toxicity was grade 5 Pneumocystis jirovecii pneumonia. Any adverse event occurred in 98.5%, grade 3/4 events in 59.0%, and grade 5 events in 10 patients. In the C2D8 aggressive-NHL subgroup (n=83), overall response was 68.7% and complete metabolic response was 56.6%. In patients without previous CAR-T exposure (n=41), the corresponding rates were 73.2% and 65.9%. 9

Evidence strength and limitation. The response rates justify interest in the combination, and pharmacodynamic changes support the proposed co-stimulatory mechanism. But there is no randomized comparator, the efficacy results come from selected dose-timing subgroups, and the safety profile includes fatal events. Roche sponsored the study and its employees participated in design, conduct, data handling, analysis, interpretation, and manuscript preparation. 11

Clinical implication. This is a development signal for patients with relapsed or refractory B-cell NHL, not a new standard of care. The next decision point is whether the apparent response gain survives a controlled comparison with a clear account of infection risk and prior CAR-T exposure.

Author affiliation and funding. First author Martin Hutchings is affiliated with Rigshospitalet and the University of Copenhagen. The study was sponsored by F. Hoffmann-La Roche Ltd. 11

Paper link: PubMed record · Nature Medicine full article

What the paper did. RAG-17 is an siRNA-oligonucleotide conjugate designed to lower SOD1 in the central nervous system. After rodent and cynomolgus-monkey work, the first-in-human dose-escalation study enrolled six patients with SOD1-ALS. Three received an initial 60 mg dose and three an initial 90 mg dose, with maintenance doses of 150 mg in five patients and 180 mg in one. The PubMed record was dated July 18, 2026. 10

Primary result. The primary safety endpoint was met. Treatment-emergent adverse events occurred in 2/6 patients (33%), all mild to moderate and resolved; no serious adverse events were reported. CSF SOD1 fell by 69% in cohort 1 at day 240 and 56% in cohort 2 at day 210. Plasma neurofilament light chain fell by 62% and 52%, respectively. No participant required invasive mechanical ventilation or died during the study. 10

Evidence strength and limitation. The paper shows durable target engagement and a tolerability signal in a genetically defined ALS subgroup. Six participants, no comparator, and no meaningful change in ALSFRS-R mean that the biomarker reductions cannot establish motor benefit or survival benefit. The preclinical results should not be treated as clinical efficacy evidence.

Clinical implication. RAG-17 supports continued testing of CNS-delivered SOD1 silencing, with later trials needing larger samples, a comparator, and functional endpoints. For clinicians, the immediate relevance is trial awareness rather than off-study treatment selection.

Author affiliation and funding. The retrieved PubMed and Crossref records did not expose a reliable first-author affiliation. Funding was listed as the Beijing Municipal Science and Technology Commission and the Beijing Nova Program. 10

Open OPTIMA-AF first if your work touches AF-PCI antithrombotic protocols: it has the clearest randomized practice question and the cleanest bleeding signal. EMPIRICAL is the paper for pediatric HIV and CMV strategy, but its early hazard finding needs to be weighed against neutral prespecified mortality comparisons. The p-tau217 paper is the strongest biomarker-methods read, with useful absolute-risk framing and an explicit warning against premature individual prediction.

Englumafusp plus glofitamab and RAG-17 are development papers. Both show biological or clinical signals, but their phase 1 designs, small or selected populations, and safety or efficacy uncertainty make them papers to read for what comes next, not for immediate protocol replacement.