Five papers: June 23, 2026

Five papers from the June 23, 2026 window — Cell delivers a mechanistic tour de force linking cuproptosis to anti-tumor immunity; JAMA Internal Medicine publishes a APOLLO consortium cohort showing APOL1 high-risk genotype more than doubles kidney function decline after living donation; a second JAMA Internal Medicine paper shows a multilevel VA stewardship program shaving antiplatelet use in DOAC patients across 135 health systems; Nature Cancer breaks a seven-day drought with a female-specific GABA-glioblastoma axis; and Nature Communications reports a phase I KRAS peptide vaccine meeting both safety and immunogenicity endpoints in MMR-proficient metastatic colorectal cancer.

Journal: Cell — IF ~67 (Tier 1) 1

Study design: Preclinical mechanistic study. Multiple immunocompetent and immunodeficient mouse tumor models; in vitro co-culture of CD8+ T cells with tumor cell lines; genetic manipulation of FDX1 (ferredoxin 1) and STAT1-IRF1 signaling; combination treatment experiments pairing cuproptosis inducers with anti-PD-L1 antibodies. Published under CC BY (open access).

Key findings: CD8+ T cell-derived IFN-γ binds its receptor on tumor cells and activates the STAT1-IRF1 transcription axis, which drives upregulation of FDX1 — the rate-limiting enzyme in the copper-dependent cell death pathway known as cuproptosis. This sensitizes tumor cells to copper-induced death. 1 The directionality is bidirectional: cuproptotic tumor cells release damage-associated molecular patterns (DAMPs) that activate dendritic cells, qualifying cuproptosis as immunogenic cell death (ICD) capable of amplifying the CD8+ T cell response. Cuproptosis inducers showed markedly more potent tumor-suppressive activity in immunocompetent hosts than in immunodeficient animals — confirming that the immune microenvironment is an active participant, not a bystander. The combination of a cuproptosis inducer plus anti-PD-L1 amplified intra-tumoral cuproptosis and overcame PD-L1 therapy resistance across multiple preclinical tumor models. 1 No competing financial interests were declared.

Boyi Gan, the paper's senior author and Hellen Hawkins Distinguished Professor and Director of the Radiation and Cancer Metabolism Program at MD Anderson, described the finding on OncoDaily: "We uncover a previously unrecognized connection between cuproptosis (copper-induced cell death) and anti-tumor immunity. We demonstrate that activating cuproptosis can engage the immune system and overcome resistance to immune checkpoint blockade, revealing a potential therapeutic strategy for immunotherapy-refractory cancers." 2

Clinical/research implications: Resistance to PD-1/PD-L1 checkpoint blockade affects roughly half of patients who receive it, and identifying mechanistic routes around that resistance is a central problem in immuno-oncology. This paper proposes that cuproptosis — a relatively recently characterized cell death modality dependent on mitochondrial lipoylation disruption by copper — is not an immunologically inert form of death but an actively immune-stimulating one. The IFN-γ → STAT1-IRF1 → FDX1 axis provides a concrete molecular explanation for why immune-hot tumors might be more susceptible to copper ionophore-type agents. Whether this translates clinically depends on the availability of safe copper-delivery agents; several copper ionophores (e.g., elesclomol) are already in clinical development for other indications. The CD8+ T cell → cuproptosis → DC activation loop suggests that cuproptosis inducers might function as in situ vaccination agents when combined with checkpoint blockade.

Limitations: The study is entirely preclinical. No patient-derived tumor data or pharmacokinetic modeling for clinical copper dosing are provided.

Authors/institution: Guang Lei, Lu Lu, Boyi Gan et al. (25+ authors); MD Anderson Cancer Center and collaborating institutions. Published June 22, 2026.

Links: PubMed PMID 42330950

Journal: JAMA Internal Medicine — IF ~39 (Tier 1) 3

Study design: Retrospective cohort study. 445 Black and 208 White US living kidney donors enrolled from the Scientific Registry of Transplant Recipients (donation years 2000–2008), with home-based research visits conducted at a median of 18.5 years post-donation. APOL1 genotyping performed on all Black donors; high-risk genotype defined as G1/G1, G2/G2, or G1/G2 compound heterozygote. DOI: 10.1001/jamainternmed.2026.0996.

Key findings: 68 of 445 Black donors (15.3%) carried APOL1 high-risk genotypes. Overall, 46 of 653 participants (7.0%) had developed eGFR <45 mL/min/1.73m² at a median of 18.5 years after donation. Black donors carrying APOL1 high-risk genotypes had a 2.31-fold higher risk of eGFR <45 compared with Black donors without high-risk genotypes (RR 2.31; 95% CI 1.16–4.61; P = .02). 3 After adjustment for predonation eGFR, the relative risk attenuated to 1.91 (95% CI 0.90–4.03; P = .09), suggesting that predonation kidney function partly mediates — but does not fully explain — the genotype-outcome relationship. The authors recommend APOL1 genotyping for all Black living kidney donor candidates in the United States.

Clinical/research implications: Living kidney donation is generally safe, but donors cede approximately half their nephron mass and absorb a lifetime residual risk of chronic kidney disease. APOL1 variants G1 and G2 arose under positive selection in sub-Saharan Africa (conferring protection against Trypanosoma brucei rhodesiense) and are carried by roughly 13–17% of Black Americans. This is the first study to quantify the interaction of APOL1 high-risk genotype with the specific physiological stress of uninephrectomy in a large prospective-recruitment cohort followed nearly two decades. The unadjusted RR of 2.31 is a clinically meaningful signal, and the attenuated adjusted RR of 1.91 (borderline non-significant with P = .09) likely reflects limited statistical power rather than absence of an effect — the confidence interval still excludes the null at the lower bound only after full adjustment, and the original RR's 95% CI is solidly significant. The implication for transplant centers is concrete: pre-donation APOL1 genotyping can identify a higher-risk subgroup among Black candidates who can then receive more individualized counseling and long-term follow-up protocols.

Limitations: Retrospective enrollment from the SRTR captures donors who were initially cleared; pre-donation selection bias toward healthier APOL1 high-risk carriers means the true population-level RR could be higher. Sample size for the APOL1 high-risk subgroup (n=68) limits statistical precision.

Authors/institution: Meyeon Park (lead author, UCSF) and the APOLLO Consortium. Published June 22, 2026.

Links: PubMed PMID 42329639 · DOI 10.1001/jamainternmed.2026.0996

Journal: JAMA Internal Medicine — IF ~39 (Tier 1) 4

Study design: Quality improvement study with interrupted time-series analysis. 7 VHA intervention sites (27,588 patients) vs. 128 VHA control sites (253,085 patients), July 2020–July 2023. Two-stage rollout: Stage 1 (9 months) — clinician and patient education plus electronic health record (EHR) modifications; Stage 2 (16 months) — addition of a pharmacist-facing electronic flag. Primary outcome: antiplatelet use rate in patients receiving direct oral anticoagulants (DOACs). DOI: 10.1001/jamainternmed.2026.2036.

Key findings: Preintervention antiplatelet use in DOAC patients was 26.1% at intervention sites and 30.1% at control sites. The combined intervention reduced antiplatelet use by −0.58 percentage points (pp) per 6 months (95% CI −0.95 to −0.22) relative to controls. 4 Stage 1 (education + EHR changes) contributed −0.29 pp per 6 months (95% CI −0.61 to 0.04) and Stage 2 (pharmacist flag) added an incremental −0.29 pp per 6 months (95% CI −0.61 to 0.03) — the two stages were additive and roughly equal in magnitude, with each stage's individual confidence interval just crossing zero. The largest effect was in stable coronary artery disease patients: −2.1 pp per 6 months (95% CI −3.0 to −1.2), equivalent to a −5.5% additional change relative to baseline prevalence. 4

Clinical/research implications: Concurrent antiplatelet and DOAC therapy is guideline-indicated in a narrow set of high-thrombotic-risk patients (e.g., recent ACS with a stent), but observational data consistently show 20–30% of DOAC-treated patients carry an antiplatelet without a guideline-supported indication — a prescription pattern that increases major bleeding risk without a commensurate thrombotic benefit. This study is notable for its scale (280,000+ patients across 135 sites), its use of the VHA's integrated EHR infrastructure to implement without requiring individual prescriber buy-in, and its additive two-stage design that allows attribution of effect to specific intervention components. The individual-stage confidence intervals crossing zero should be interpreted carefully: the interrupted time-series design has limited power to detect small per-period changes in a short window, and the composite effect — which is the operationally relevant metric for scaling — is significant. The stable CAD subgroup effect (−2.1 pp/6 months) is the most practically meaningful finding, given that stable CAD patients are the largest inappropriate-antiplatelet subgroup in DOAC registries.

Limitations: Non-randomized quality improvement design; contamination between adjacent VHA sites cannot be fully excluded. The 16-month Stage 2 window may be too short to observe the full pharmacist-flag effect.

Authors/institution: Jacob Kurlander (lead, VA Ann Arbor Healthcare System) and Jeremy Sussman (senior, VA Ann Arbor / University of Michigan). Published June 22, 2026.

Links: PubMed PMID 42329643 · DOI 10.1001/jamainternmed.2026.2036

Journal: Nature Cancer — IF ~23 (Tier 1) 5

Study design: Preclinical mechanistic study plus human translational validation. Syngeneic GBM mouse models (SB28 and GL261); GABBR agonist (baclofen) and antagonist (CGP 35348) dosing experiments; pharmacological NOS2 inhibition (aminoguanidine); conditional CAT2-knockout mice; anti-Ly6G granulocytic MDSC (gMDSC) depletion; L-arginine metabolite profiling; human GBM tumor immune cell transcriptomics; human peripheral blood mononuclear cell (PBMC) baclofen stimulation. Multi-institutional: University of Miami, University of Michigan, Cleveland Clinic, Florida International University, Vanderbilt University. Corresponding author: Defne Bayik (University of Miami).

Key findings: GABA (γ-aminobutyric acid), the brain's predominant inhibitory neurotransmitter, activates GABA-B receptor (GABBR) signaling on gMDSCs within the GBM tumor microenvironment. This upregulates the amino acid transporter CAT2B (encoded by Slc7a2), driving L-arginine accumulation, NOS2 (nitric oxide synthase 2) activation, and peroxynitrite production — a potent T cell-suppressive mediator. The pathway is female-specific: baclofen (GABBR agonist) significantly accelerated GBM progression in female mice (SB28 model: P = 0.001; GL261 model: P < 0.001) but had no effect in male mice or in immunodeficient animals. 5 GABBR antagonist CGP 35348 extended survival specifically in female mice (P = 0.031) and reduced NOS2 expression in tumor-infiltrating gMDSCs — without affecting male mice. CAT2-knockout mice were unresponsive to baclofen's tumor-promoting effect; anti-Ly6G depletion of gMDSCs also rescued baclofen-treated females, confirming gMDSC dependence. GABA had no direct effect on GBM cell proliferation in vitro, indicating the effect is entirely microenvironment-mediated. Human validation: female GBM patient immune cells showed enriched GABA transcriptional signatures; GABA concentration was higher in female tumors (P = 0.018 in mice; human data from 8 high-grade glioma samples show the same direction). In human PBMCs, baclofen increased NOS2 in female gMDSCs (P = 0.029) but not male. Baseline L-arginine levels and Slc7a2 expression are lower in female gMDSCs at baseline, which explains why the GABA-induced upregulation is amplified in females.

Clinical/research implications: GBM is uniformly fatal, with a median survival of 14–16 months from diagnosis and limited benefit from second-line therapies. Clinical immunotherapy trials in GBM have largely failed, and the reasons for this — relative to the partial successes seen in other cancers — remain incompletely understood. This paper offers a specific, sex-stratified mechanistic explanation for why gMDSC-mediated T cell suppression is more pronounced in female GBM patients, which is notable given that female sex is historically associated with better GBM prognosis — the biology appears to be more complex than a simple survival advantage. Baclofen is an approved drug for spasticity; the finding that it accelerates GBM in females argues for caution in its use in female GBM patients and invites retrospective pharmacoepidemiological analysis of baclofen exposure in existing GBM cohorts. GABBR antagonism or downstream NOS2 inhibition represents a therapeutically accessible node, as both the receptor and the pathway have clinically characterized pharmacology. Sex-stratified enrollment and analysis in future GBM immunotherapy trials is a direct design implication.

Limitations: Small human tumor sample sizes (n=2 female, n=6 male high-grade glioma for the GABA concentration analysis). Mouse models may not fully recapitulate human GBM immune architecture. The 22-month review period suggests the paper underwent substantial revision.

Authors/institution: Pathak, Bayik et al. (38 authors); University of Miami (corresponding: Defne Bayik). Received July 23, 2024; accepted May 22, 2026. Published June 23, 2026.

Links: Nature Cancer DOI 10.1038/s43018-026-01192-5

Journal: Nature Communications — IF ~16 (Tier 2, open access) 6

Study design: Single-arm phase I trial (NCT04117087). 13 patients with pretreated metastatic MMR-proficient (MMRp) / microsatellite-stable (MSS) colorectal cancer — a population in which single-agent PD-1 blockade has essentially no activity. Intervention: mKRAS-VAX (a pooled synthetic peptide vaccine targeting 6 KRAS driver mutations: G12A, G12C, G12D, G12R, G12V, G13D) combined with nivolumab (anti-PD-1, Bristol Myers Squibb) and ipilimumab (anti-CTLA-4, BMS), plus poly-ICLC (Hiltonol®, Oncovir) as adjuvant. Two co-primary endpoints: safety and immunogenicity. Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University. Corresponding authors: Nilofer S. Azad and Neeha Zaidi. Received December 2, 2025; accepted June 8, 2026. Funded by NIH (R01CA296410, R01CA197296, P50CA062924, and others). COI: Elizabeth M. Jaffee is a founder of Adventris Pharmaceuticals, which holds a license on the vaccine technology.

Key findings: Both co-primary endpoints were met. All vaccine-attributed adverse events were grade 1–2; the combination did not increase severe immune-related AEs beyond the known profile of dual ICI (nivolumab + ipilimumab) alone. 6 For immunogenicity: tumor-specific mKRAS-reactive T cells were detected in 8 of 12 biomarker-evaluable patients (67%) by direct ex vivo IFN-γ ELISpot within 17 weeks of vaccination; after in vitro expansion, all 12 of 12 patients (100%) showed a T cell response. 6 The authors conclude: "Our findings support further development of mKRAS vaccines with ICIs for advanced MMRp/MSS CRC."

Clinical/research implications: MMRp/MSS metastatic CRC accounts for approximately 85% of metastatic colorectal cancer cases and has a dismal prognosis once standard chemotherapy and VEGF/EGFR-directed therapies are exhausted. The failure of PD-1 blockade in this histotype — in contrast to the dramatic responses seen in MSI-high CRC — has been attributed to the absence of a pre-existing T cell response (low tumor mutational burden, immune-cold microenvironment). A therapeutic cancer vaccine targeting the mutant-KRAS neoantigen is conceptually designed to prime that T cell response de novo. This phase I trial is too small (n=13) and uncontrolled to provide efficacy conclusions; the immunogenicity data — particularly the 67% direct ex vivo response rate in a historically immunologically cold tumor type — are the signal of interest, and they support advancing to a dose-expansion or randomized phase. The 100% in-vitro expansion response rate confirms the mKRAS antigen is immunogenic but should be interpreted as a pharmacodynamic assay result rather than a clinical response predictor. Whether vaccine-induced T cells can penetrate and functionally persist in the MMRp/MSS CRC tumor microenvironment is the critical translational question for phase II design.

Limitations: Single-arm, 13 patients; no efficacy endpoint reported in this analysis; all-grade AEs only from the vaccine component; the COI structure (Jaffee / Adventris licensing) is standard for academic translational trials but should be noted.

Authors/institution: Wang, Huff, Zaidi, Azad et al. (multiple authors); Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University. Published June 23, 2026.

Links: Nature Communications DOI 10.1038/s41467-026-74711-8 · ClinicalTrials.gov NCT04117087