May 2025 — Monthly Briefing¶
1. What was read¶
812 research articles from 33 of the 33 journals on the roster.
A further 276 records (25% of the 1,088 this month's harvest returned) carry no abstract and are not research: news, correspondence, editor highlights and obituaries. PubMed types them Journal Article and nothing else, so the publication-type filter cannot see them — across the whole index they are 22.7% of it and 35% of general, and 3,801 of them are JAMA. Every count below is over the full harvest and is inflated by this; the figure above is not.
| General | Cancer | Immune | Bioinfo | Sysbio |
|---|---|---|---|---|
| 538 | 220 | 88 | 222 | 20 |
Largest contributors: Nature 122, JAMA 118, Science 108, J Immunother Cancer 59, Nucleic Acids Res 52, Cell 49.
2. Net change on the thread curve¶
| Thread | This month | Cumulative | Status |
|---|---|---|---|
agentic-ai |
— | +9 / −1 | forming |
cell-therapy |
+7 / −4 | +176 / −52 | forming |
ctdna-mrd |
+8 | +61 / −22 | forming |
foundation-models |
+1 / −2 | +93 / −64 | splitting |
immunotherapy-resistance |
+8 / −4 | +188 / −69 | forming |
interception |
+5 | +105 / −21 | forming |
method-credibility |
+1 / −11 | +14 / −385 | crisis |
neoantigen-vaccines |
+4 / −2 | +69 / −30 | forming |
neuro-immune-microbiome-axis |
+5 | +145 / −14 | forming |
next-gen-checkpoints |
+4 | +61 / −35 | splitting |
single-cell-3d-genome |
+5 | +61 / −17 | forming |
suppressive-population-inversion |
+1 / −4 | +42 / −69 | contested |
Status is derived by rule, not asserted. Fewer than 5 cards → too-early; then refutes ≥ 2×supports → crisis; refutes > supports → contested; supports ≥ 2×refutes → forming; else splitting.
3. The evidence that moved it¶
The checkpoint roster this month is TIM-3, PILRalpha/CD99, CLEVER-1, Cbl-b and CISH, plus IDO1, ENT1, OX40 and LAG-3. Of these, exactly one — CISH — has a first-in-human trial, and it works by gene editing rather than blockade. By August, zero of the five headline names appear anywhere in the corpus. The half-life measurement that the later briefings quote starts here, and this is the month it can be checked against a written list rather than a recollection.
Tumour antigens preferentially derive from unmutated genomic sequences (40405018, Nat Cancer), and pancreatic cancer-restricted cryptic antigens are targets for T cell recognition (40339010, Science). Two independent statements in one month that the immunogenic target space is not where the mutation-calling pipeline is looking. The neoantigen thread's later troubles — prediction models exploiting shortcuts, immunoediting stripping clonal targets — are downstream of this.
DYNAMIC reports five-year outcomes
(40055522): ctDNA-guided adjuvant
therapy in stage II colon cancer, randomized, mature. This is the thread's
strongest support in the corpus, and it is worth naming because almost everything
that lands on ctdna-mrd afterwards is a refutation. The distinction that
survives: ctDNA works as a de-escalation trigger, and has repeatedly failed as
an escalation trigger.
4. Refutations¶
(a) Landing on a tracked thread¶
39918337 (Cancer Discov): stromal KITL maintains pancreas homeostasis and restrains tumour progression, with a companion arguing that stellate cells uphold inherent microenvironmental tumour suppression. Pancreatic stroma depletion has been a therapeutic goal for a decade.
40316306: FMO2-positive cancer-associated fibroblasts sensitize anti-PD-1 therapy in hepatocellular carcinoma patients — human data, not mouse.
Corrected 2026-08-27
40210998 was listed alongside it — "Tregs augment self-tolerance during infection" — as a second inversion landing on the same thread in the same month. It is a three-page Nature Immunology News and Views, and its subject is 40168991, the soluble-CTLA-4 paper carded on the same thread in April with the same stance. Two cards, one study.
The result stands where April put it: effector-Treg sCTLA-4 augments self-tolerance. May adds no observation of its own. The reason the genre was invisible to the reader who carded it is that the index holds no abstract for the record, only the title — which is exactly the condition under which a commentary reads as a finding. A convergence argument that double-counts is worse than a wrong stance, because it reads as independent evidence.
(b) Method credibility¶
40346587: most human DNA replication initiation is dispersed, with only a minority inside previously annotated initiation zones. The reference annotation captures the minority case.
40193392: standard negative sampling leaves an exploitable shortcut in protein-protein interaction prediction. This is the same finding February 2026 reports for transcription-factor binding sites and May 2026 for drug synergy — the negative set, not the model, is where the performance came from. Three instances, ten months apart, in three subfields, and none of them cites the others.
40341387: bacterial pan-genome analysis approaches carry pitfalls that show up as soon as the species is not clonal.
40339003 (Science): a published Cambrian mollusc is reassigned to a different phylum-level group. Out of corpus scope, carded because the audit ledger should include the case where the original claim was simply wrong about what the object was.
5. Open items¶
Nothing carried forward.
6. Meetings¶
No abstract book published this month. The abstract layer runs 2023-02 to 2026-08 and is seasonal by nature — meetings are annual, so most months are empty here and that is the expected shape, not a gap in the harvest.
7. Regulatory¶
| Event | Count |
|---|---|
| trial-first-posted | 528 (of which 478 genuinely oncology) |
| fda-approval | 4 |
| ema-authorisation | 3 |
4 FDA actions — dated from each notice's own prose, not the table's posting date, which disagrees on 7 of 190 rows and once by enough to move an approval into the wrong month:
- 2025-05-08 — FDA grants accelerated approval to the combination of avutometinib and defactinib for KRAS-mutated recurrent low-grade serous ovarian cancer
- 2025-05-14 — FDA grants accelerated approval to telisotuzumab vedotin-tllv for NSCLC with high c-Met protein overexpression
- 2025-05-14 — FDA approves belzutifan for pheochromocytoma or paraganglioma
- 2025-05-15 — FDA approves retifanlimab-dlwr with carboplatin and paclitaxel and as a single agent for squamous cell carcinoma of the anal canal
1 EMA oncology authorisation:
- 2025-05-26 — Jubereq
8. Trade press¶
46 on-topic of 354 items across 3 feeds (n-stat 340, n-aacr-blog 11, n-acir 3). Headlines are selected by subject keyword, not recency — most of what these feeds carry is health policy and industry finance the corpus does not read.
- 2025-05-31 — STAT+: FDA approves a next-generation Covid vaccine from Moderna, with restrictions —
n-stat - 2025-05-31 — STAT+: Combination treatment reduces risk of tumor progression in aggressive form of breast cancer, study shows —
n-stat - 2025-05-30 — STAT+: Illumina’s PromoterAI unlocks rare disease clues in noncoding genome —
n-stat - 2025-05-30 — STAT+: Summit shows mixed results from key study of lung cancer drug —
n-stat - 2025-05-30 — STAT+: Next-gen COPD drug from Sanofi, Regeneron shows mixed results in Phase 3 trials —
n-stat - 2025-05-29 — AACR Annual Meeting 2025 Plenaries: Cancer Evolution, Tumor Microenvironment, Drugging KRAS, and More —
n-aacr-blog - 2025-05-28 — Navigating the Meandering Path of Oncology Drug Development —
n-aacr-blog - 2025-05-27 — From the Bench, Spring 2025: Making Tumors Pig-like, Transplanting Fat Cells, and Other Creative Approaches in Cancer Research —
n-aacr-blog - 2025-05-27 — STAT+: Pharmalittle: We’re reading about a Roche antibiotic, a gene therapy patient death, and more —
n-stat - 2025-05-27 — STAT+: Participant in Rocket Pharmaceuticals gene therapy study dies —
n-stat - 2025-05-27 — STAT+: At a major genome-editing summit, spotlight turns to the value of human life —
n-stat - 2025-05-26 — STAT+: What Baby KJ means for the CRISPR gene-editing industry —
n-stat - … and 34 more on-topic items.