February 2024 — Monthly Briefing¶

1. What was read¶

736 research articles from 32 of the 33 journals on the roster.

A further 212 records (22% of the 948 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
484	139	77	226	22

Volume swing against the median month: cancer -25%

Check the source before reading anything into this. Two benign causes account for most of these. bioinfo every January carries the Nucleic Acids Research database issue and runs high. And several journals simply deposit nothing in some months: across the 44 months read, 4 of 33 journals are absent from at least one, worst being Nat Mach Intell (20 months missing, 81 papers in total) and Bioinformatics (4 months missing, 2,905 papers in total). This month is missing Nat Mach Intell. Queried directly, PubMed returns zero for those journal-months too — the holes are in the source, reproduced exactly by re-running the harvest. So a swing here is not evidence that a field went quiet, and is not usually evidence that the pipeline broke either.

Largest contributors: Nature 114, Science 89, JAMA 82, Nucleic Acids Res 77, Bioinformatics 60, Cell 49.

2. Net change on the thread curve¶

Thread	This month	Cumulative	Status
`agentic-ai`	+1	+4 / −0	too-early
`cell-therapy`	+8 / −3	+94 / −30	forming
`ctdna-mrd`	+5	+30 / −13	forming splitting → forming
`foundation-models`	+2 / −2	+36 / −25	splitting
`immunotherapy-resistance`	−1	+101 / −35	forming
`interception`	—	+61 / −9	forming
`method-credibility`	+1 / −14	+3 / −172	crisis
`neoantigen-vaccines`	+3	+42 / −17	forming
`neuro-immune-microbiome-axis`	+1	+64 / −8	forming
`next-gen-checkpoints`	+2 / −1	+27 / −19	splitting
`single-cell-3d-genome`	—	+26 / −7	forming
`suppressive-population-inversion`	—	+23 / −30	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 vaccine thread's best clinical evidence is from February 2024. 38246194 (Lancet) is KEYNOTE-942: individualised neoantigen therapy mRNA-4157 plus pembrolizumab versus pembrolizumab alone in resected melanoma, randomised. Nothing later in this corpus supports the neoantigen thread as strongly, and by 2026-07 the same thread is being attacked on both its target space and its selection method. The high-water mark is at the start.

The regulator states its own case in NEJM, and the field answers that the risk is small. 38265704 is a perspective by Verdun and Marks of CBER on secondary cancers after CAR-T: the FDA's own account of the action it has taken, in the flagship clinical journal, over the names of the two officials running the review. 38206604 carries the safety notice in JAMA. Against them 38195751 (Nat Med) reviews the reports of T cell malignancies and concludes that existing follow-up data suggest a low risk compared with other cancer treatments — a reassurance, not a warning, and the safety signal that would have constrained the platform is on this reading not borne out.

Corrected 2026-08-27

This paragraph was headed "CAR-T's secondary-malignancy problem reaches NEJM in one month" and argued that "January's regulatory signal becomes February's flagship clinical literature — a four-week lag, which is the fastest this corpus records for anything". Both halves fail. 38265704 has no abstract because it is not a study: it is a three-page perspective, pages 584-586, by two FDA officials, so the February item is the same regulatory action as December's and January's, not the literature responding to it. There is no lag between them because there are not two things. And 38195751 was carded as a refutation on the strength of "unanswered questions" in its title; its one-sentence abstract reaches the opposite conclusion, and its stance is now supports. A safety-reassurance commentary is in any case a weak fit for a thread whose claim is about the platform generalising beyond CD19; a later reader should feel free to drop it rather than inherit it.

And the indication that saves the platform is already in NEJM. 38381673: CD19 CAR-T in autoimmune disease, a case series with follow-up. The corpus's November 2025 briefing calls autoimmunity CAR-T's growth edge and treats it as a recent turn. It is in the flagship clinical journal in February 2024.

ctDNA is measured well and acted on nowhere. 37972346 prospectively validates HPV ctDNA after chemoradiation in cervical cancer: across 70 patients, persistent ctDNA is independently associated with inferior two-year progression-free survival — 82% against 24% for those still positive at three months — with a median lead time to recurrence of 5.9 months. 37992871 pairs ctDNA detectability with CT-derived tumour volume in 85 resected NSCLC patients and a 32-patient validation cohort, and stratifies relapse risk. Both are clean prognostic associations; neither is a treatment decision, and the first says so in its own conclusion, which offers the assay to future intensification trials. ctdna-mrd claims the analyte guides treatment and does not merely forecast it, and all three of this month's ctDNA cards are neutral.

Corrected 2026-08-27

This paragraph was headed "ctDNA converges on residual disease" and closed "Detection is being solved here." Two of its three citations were carded supports for the thread defined against exactly the use they make of the assay. The third, 38361060, was described as attacking "the sensitivity floor at the assay level"; it is not a method paper at all, but a single-page Nature Biotechnology research highlight on page 184, written by a journal editor, with no abstract and no data, summarising work published elsewhere. Journal front matter cannot be evidence for the thread the journal is reporting on.

4. Refutations¶

(a) Landing on a tracked thread¶

The FDA perspective above is the month's refutation on cell-therapy and is filed as one; the Nature Medicine commentary beside it is not, and now reads as support. It is worth reading both against the same month's 38316518, a complete response in gastric cancer from Claudin18.2 CAR-T — the target that in June 2026 becomes the first CAR-T approved anywhere for a solid tumour. Both the platform's liability and its eventual solid-tumour success are visible in the same month.

(b) Method credibility¶

38200587 (Bioinformatics): pitfalls of machine learning models for protein-protein interaction networks. The paper's finding is that the field had no consistent evaluation framework, so its networks were not comparable and the discrepancies between algorithms went unexplained; the benchmarking framework it builds to fix that then shows algorithm design has little impact on performance once functional-genomics features are in play. That is a credibility result about how the models were being compared, and it replicates in human and yeast data.

Corrected 2026-08-27

This entry read: "This is the direct ancestor of the July 2025 finding that sequence-based interaction predictors plateau at 0.65 — the same objection, seventeen months earlier, in the same journal." There is no plateau and no such number in the abstract. What it reports about sequence is the opposite shape: sequence-based and functional-genomics models are complementary, the first specialising in interactions involving hub proteins and the second doing better on lone ones. A later result was written backwards onto an earlier paper. The card has also lost foundation-models; these are classic supervised models, with nothing pretrained or general-purpose in the paper.

38331411: zero is not absence. The standard handling of zeros in microbiome differential-abundance testing is wrong and censoring-based analysis is required. 38279041 (Nat Genet): adjusting for genetic confounders in transcriptome-wide association studies improves discovery — which says the unadjusted analyses already in the literature were confounded.

5. Open items¶

Nothing carried forward.

6. Meetings¶

708 abstracts from 1 meeting.

Meeting	Abstracts	With full text
ASCO Annual Meeting 2024	708	708

7. Regulatory¶

Event	Count
trial-first-posted	555 (of which 500 genuinely oncology)
ema-authorisation	7
fda-approval	4

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:

2024-02-13 — FDA approves irinotecan liposome for first-line treatment of metastatic pancreatic adenocarcinoma
2024-02-15 — FDA approves tepotinib for metastatic non-small cell lung cancer
2024-02-16 — FDA approves osimertinib with chemotherapy for EGFR-mutated non-small cell lung cancer
2024-02-16 — FDA grants accelerated approval to lifileucel for unresectable or metastatic melanoma

2 EMA oncology authorisations:

2024-02-09 — Mevlyq
2024-02-16 — Pomalidomide Viatris

8. Trade press¶

31 on-topic of 273 items across 3 feeds (n-stat 260, n-aacr-blog 9, n-acir 4). 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.

2024-02-29 — STAT+: Icon acquires HumanFirst to boost support for digital tools in clinical trials — n-stat
2024-02-29 — STAT+: Pfizer is about to make a big push in cancer. Will investors listen? — n-stat
2024-02-28 — MEGA news in gene editing: using new CRISPR technology for RNA knockdown — n-acir
2024-02-22 — STAT+: FDA gives a mixed response to a petition seeking greater clinical trial transparency — n-stat
2024-02-22 — Humira’s legacy, CEO symbolism, and genomic surgery — n-stat
2024-02-22 — War zone or not, Ukraine seeks to reclaim its role as a hub for clinical trials — n-stat
2024-02-21 — Dragging Oncogenes From the Cell: A New Horizon for Targeting KRAS — n-aacr-blog
2024-02-21 — Small study adds to growing hope CAR-T cell therapy could revolutionize autoimmune disease treatment — n-stat
2024-02-21 — Meet the fetal surgeon forging CRISPR’s next frontier: curing diseases in the womb — n-stat
2024-02-21 — Taking tips from T cell cancers to improve T cell therapies — n-acir
2024-02-20 — STAT+: Ring Therapeutics, a Flagship gene therapy startup, lays off 20% of staff — n-stat
2024-02-20 — STAT+: Pharmalittle: We’re reading about weight loss drugs, a cell therapy for melanoma, and more — n-stat
… and 19 more on-topic items.