Skip to main content
Tessa NewsLink
Paste a health news link, or browse

Source study found

Story checked

Screening method identifies leukemia drug candidates from more than 4 million compounds (opens in a new tab)

medicalxpress.com · 2026-09-18

Short answerEvidenceSource

Short answer

Mixed

Mixed.

The claims we could check match the study, but some claims were not covered by the evidence reviewed.

  • 2 supported
  • 3 not covered

Checked against the study summary. The full text wasn't available, so some details couldn't be settled either way.

Share this check

Follow the evidence trail
1
2

NewsLink checks it

Mixed

Every claim we could check holds up. Two of five claims match the study. This overall rating is based only on the claims we could check. Three claims the study doesn't address.

  • 2 supported
  • 3 not covered
Open claim evidence
3
Then inspect each claim

Evidence layer

Claim by claim

Each claim gets a verdict. Expand it to see the evidence directly below.

5 claims in this story

Showing all 5 claimsChoose a verdict to focus the list.

Claim 1 of 5Not coveredThe team used machine learning to work backward from prior wet-lab screens and predicted compounds from more than 4 million drug compounds, with fewer than 100 meeting the selection criteria.View evidenceHide evidence

As statedmore than 4 million; fewer than 100

Why this verdict

The abstract-level profile supports screening approximately 4.2 million compounds and prioritizing compounds based on predicted tri-functional activity. However, the supplied paper profile does not verify the story's more specific claims that this was explicitly 'machine learning' working backward from prior wet-lab screens, nor the stated magnitude that fewer than 100 compounds met selection criteria. The profile explicitly notes that number of hits and hit-selection thresholds are not reported in the abstract.

1

Study evidence

A tri-functional cheminformatic screen of ~4.2 million compounds identified hits that, upon experimental validation, selectively killed AML cells, supporting the prediction-driven discovery approach.

“Using a cheminformatic screen of ~4.2 million compounds for molecules with high predicted probability for all three functions, we found and validated hits that selectively killed AML cells”
Claim 2 of 5Not coveredWhen the best compounds were tested in AML cell lines and patient-derived AML cells, the article says they selectively targeted AML cells and showed synergy with existing AML drugs.View evidenceHide evidence

As statedup to 95%–97% of the cancer cells but only 5% of healthy blood cells

Why this verdict

The qualitative parts are supported at abstract level: identified compounds selectively killed AML cells in vitro, and drug-combination synergy was reported with established AML therapies, with some confirmation in AML patient-derived primary cells. However, the story's specific quantitative claim—killing up to 95%–97% of cancer cells but only 5% of healthy blood cells—is not present in the abstract-level profile, nor are the exact experimental contexts and quantitative selectivity metrics available. Therefore the full presented claim, including the stated magnitude, is not verifiable at this evidence depth.

1

Study evidence

Cheminformatic-identified hits selectively killed AML cells in vitro.

“validated hits that selectively killed AML cells, activated apoptosis, were dependent upon autophagic activation”
2

Study evidence

The identified compounds showed strong in vitro synergy with midostaurin and venetoclax in AML cells; synergy with doxorubicin and midostaurin was confirmed in patient-derived AML primary cells.

“we also observed strong synergy between these compounds and midostaurin and venetoclax”
Claim 3 of 5Not coveredThe article says inhibiting glutathione reductase exploits a leukemia-cell vulnerability related to high mitochondrial activity and reactive oxygen species (ROS) overload.View evidenceHide evidence

Why this verdict

The abstract-level profile supports that compound treatment interfered with glutathione reductase/glutathione metabolism, increased cytosolic and mitochondrial ROS, and reduced oxygen consumption and ATP synthesis in AML cells. But the broader causal interpretation that glutathione reductase inhibition exploits a leukemia-cell vulnerability specifically tied to high mitochondrial activity and ROS overload is not fully established from the abstract-level evidence supplied. The profile also notes that direct mechanism claims about glutathione reductase inhibition versus indirect effects cannot be fully evaluated from the abstract alone.

1

Study evidence

Compound treatment interfered with glutathione reductase and compromised glutathione metabolism in AML cells.

“compromised glutathione metabolism by interfering with glutathione reductase”
2

Study evidence

The compounds' impact on glutathione metabolism and redox balance was confirmed in AML patient-derived primary cells.

“Key phenotypes, including the compounds' impact on glutathione metabolism and synergy with doxorubicin and midostaurin, were confirmed in AML-patient-derived primary cells”
Claim 4 of 5SupportedA paper recently published in Leukemia described a new screening method that identified drug candidates for acute myeloid leukemia.View evidenceHide evidence

Why this verdict

The abstract-level profile supports that the paper reported a new prediction-driven cheminformatic/virtual screening approach that identified AML drug-candidate hits and that those hits were experimentally validated for selective AML cell killing. The story frames these as candidates/potential drugs rather than established clinical treatments, which matches the abstract-level evidence.

1

Study evidence

A tri-functional cheminformatic screen of ~4.2 million compounds identified hits that, upon experimental validation, selectively killed AML cells, supporting the prediction-driven discovery approach.

“Using a cheminformatic screen of ~4.2 million compounds for molecules with high predicted probability for all three functions, we found and validated hits that selectively killed AML cells”
2

Study evidence

Cheminformatic-identified hits selectively killed AML cells in vitro.

“validated hits that selectively killed AML cells, activated apoptosis, were dependent upon autophagic activation”
Claim 5 of 5SupportedThe screening focused on compounds that increased apoptosis, increased autophagy, and inhibited glutathione reductase.View evidenceHide evidence

Why this verdict

The abstract-level profile supports that the screen selected for three predicted functions: apoptotic agonism, thioredoxin/glutathione reductase inhibition, and autophagy induction. The story's wording of increased apoptosis, increased autophagy, and inhibited glutathione reductase is a fair reflection, though the paper profile includes thioredoxin/glutathione reductase rather than glutathione reductase alone.

1

Study evidence

A tri-functional cheminformatic screen of ~4.2 million compounds identified hits that, upon experimental validation, selectively killed AML cells, supporting the prediction-driven discovery approach.

“Using a cheminformatic screen of ~4.2 million compounds for molecules with high predicted probability for all three functions, we found and validated hits that selectively killed AML cells”
2

Study evidence

Cheminformatic-identified hits selectively killed AML cells in vitro.

“validated hits that selectively killed AML cells, activated apoptosis, were dependent upon autophagic activation”
Then look for missing context

Context layer

What the story left out

Important study details the story did not include.

  • Structurally unrelated compounds recapitulated the same cellular phenotypes, supporting screening by predicted function rather than shared chemical scaffold.

    The supplied story presentation does not mention the cross-scaffold replication/generalizability test, which is a material abstract-level element supporting the screening approach.

    From comparative phenotyping across chemotypes (in vitro)

5 things the story did carry across
  • Large-scale tri-functional cheminformatic screen of approximately 4.2 million compounds for apoptotic agonism, thioredoxin/glutathione reductase inhibition, and autophagy induction.
  • Experimental validation that prioritized hits selectively killed AML cells and activated apoptosis in vitro, with autophagy dependence reported.
  • Mechanistic redox and mitochondrial phenotypes: glutathione reductase/glutathione metabolism interference, increased cytosolic and mitochondrial ROS, decreased oxygen consumption, and reduced ATP synthesis.
  • Drug-combination synergy with established AML therapies, including midostaurin and venetoclax in AML cells, and doxorubicin/midostaurin in patient-derived primary cells.
  • Ex vivo confirmation in AML patient-derived primary cells of selected key phenotypes, including glutathione metabolism effects and drug synergy.
Then read the study layer

Study layer

Study at a glance

Scan the study first. Expand only the parts you want to inspect.

Pieces of work

6

Evidence read

study summary

Lead result

in silico

1Lead resultin silicoA cheminformatic screen for compounds with tri-functional predicted activity (apoptotic agonism, thioredoxin/glutathione reductase inhibition, and autophagy induction) yields hits that selectively kill AML cells and validate the prediction-driven discovery approach.cheminformatic virtual screenExpand

In plain English

The authors performed a large-scale cheminformatic/virtual screen of ~4.2 million compounds to identify molecules with high predicted probability of three functions—apoptotic agonism, thioredoxin/glutathione reductase inhibition, and autophagy induction—and prioritized hits for experimental follow-up. Hits from this tri-functional prediction screen were experimentally validated to selectively kill AML cells, supporting the prediction-driven discovery approach.

Key findings

  • A tri-functional cheminformatic screen of ~4.2 million compounds identified hits that, upon experimental validation, selectively killed AML cells, supporting the prediction-driven discovery approach.
“Using a cheminformatic screen of ~4.2 million compounds for molecules with high predicted probability for all three functions, we found and validated hits that selectively killed AML cells”
What this piece can’t prove

3 further details could not be confirmed from the summary.

2in vitroMechanistic cellular phenotypes of the identified compounds (apoptosis activation, autophagy-dependence, glutathione reductase interference/glutathione metabolism compromise, ROS increase, mitochondrial respiration/ATP suppression) are consistent with predicted functions and are recapitulated by structurally unrelated compounds.in vitro cell-based validationExpand

In plain English

Cheminformatic hits (PS127-family and additional compounds) were validated in vitro: they selectively killed AML cells, activated apoptosis, and their cytotoxicity required autophagic activation. Structurally unrelated compounds produced the same cellular phenotypes, supporting the predicted multimodal mechanism (apoptotic agonism, autophagy induction) in cell-based assays.

Key findings

  • Cheminformatic-identified hits selectively killed AML cells in vitro.
  • Compound treatment activated apoptosis in AML cells.
“validated hits that selectively killed AML cells, activated apoptosis, were dependent upon autophagic activation”
What this piece can’t prove
  • In vitro cell-based results may not fully predict in vivo efficacy or safety; the abstract does not provide in vivo validation for these specific cellular phenotypes.

2 further details could not be confirmed from the summary.

3in vitroMechanistic cellular phenotypes of the identified compounds (apoptosis activation, autophagy-dependence, glutathione reductase interference/glutathione metabolism compromise, ROS increase, mitochondrial respiration/ATP suppression) are consistent with predicted functions and are recapitulated by structurally unrelated compounds.in vitro mechanistic assaysExpand

In plain English

In AML cell models, selected small-molecule hits identified by cheminformatic screening interfered with glutathione reductase and compromised glutathione metabolism, increased cytosolic and mitochondrial reactive oxygen species (ROS), and decreased oxygen consumption and ATP synthesis; these mechanistic phenotypes were recapitulated by structurally unrelated compounds and key glutathione-related effects were confirmed in AML patient–derived primary cells.

Key findings

  • Compound treatment interfered with glutathione reductase and compromised glutathione metabolism in AML cells.
  • Compound treatment increased pools of cytosolic and mitochondrial reactive oxygen species (ROS).
“compromised glutathione metabolism by interfering with glutathione reductase”
What this piece can’t prove

3 further details could not be confirmed from the summary.

4in vitroMechanistic cellular phenotypes of the identified compounds (apoptosis activation, autophagy-dependence, glutathione reductase interference/glutathione metabolism compromise, ROS increase, mitochondrial respiration/ATP suppression) are consistent with predicted functions and are recapitulated by structurally unrelated compounds.comparative phenotyping across chemotypes (in vitro)Expand

In plain English

The paper reports that structurally unrelated small molecules reproduced the same AML-selective cellular phenotypes previously observed for the PS127-family compounds. Specifically, these independent compounds activated apoptosis, required autophagic activation for cytotoxicity (autophagy-dependence), interfered with glutathione reductase and compromised glutathione metabolism, increased cytosolic and mitochondrial reactive oxygen species (ROS), and reduced mitochondrial oxygen consumption and ATP synthesis. The authors present these cross-scaffold replications as validation that screening for predicted functions (apoptotic agonism, thioredoxin/glutathione reductase inhibition, and autophagy induction) — rather than shared chemical scaffold — identifies compounds with the described AML cellular effects.

Key findings

  • Structurally unrelated compounds reproduced the AML-selective cytotoxicity observed for PS127-family compounds.
  • Independent compounds activated apoptosis in AML cells consistent with predicted apoptotic agonism.
“Structurally unrelated compounds caused the same phenotypes, validating our approach of screening for predicted function.”
What this piece can’t prove

3 further details could not be confirmed from the summary.

5in vitroThe identified compounds show drug-combination synergy with established AML therapies (e.g., midostaurin, venetoclax; also doxorubicin) suggesting therapeutic potential.drug combination assays / synergy quantificationExpand

In plain English

Abstract reports that the identified small molecules show strong in vitro synergy with established AML therapies midostaurin and venetoclax in AML cells; synergy with doxorubicin and midostaurin was also confirmed in AML patient–derived primary cells. The synergy experiments were performed as drug-combination assays (dose–response matrices and synergy quantification implied in methods profile) and are presented by the authors as supporting the compounds' therapeutic potential.

Key findings

  • The identified compounds showed strong in vitro synergy with midostaurin and venetoclax in AML cells; synergy with doxorubicin and midostaurin was confirmed in patient-derived AML primary cells.
“we also observed strong synergy between these compounds and midostaurin and venetoclax”
What this piece can’t prove

2 further details could not be confirmed from the summary.

6ex vivo humanKey phenotypes (effects on glutathione metabolism and selected drug synergies) are confirmed in AML patient-derived primary cells, supporting translational relevance.ex vivo primary cell assayExpand

In plain English

Ex vivo testing in AML patient-derived primary cells confirmed key mechanistic and combination-treatment phenotypes of cheminformatically identified compounds: they disrupted glutathione metabolism/redox balance and showed synergistic activity with selected chemotherapeutics (doxorubicin) and targeted therapy (midostaurin). These observations were reported as validation of translational relevance in primary human AML specimens.

Key findings

  • The compounds' impact on glutathione metabolism and redox balance was confirmed in AML patient-derived primary cells.
  • Ex vivo combination testing in AML primary cells showed synergy between the compounds and doxorubicin, and between the compounds and midostaurin.
“Key phenotypes, including the compounds' impact on glutathione metabolism and synergy with doxorubicin and midostaurin, were confirmed in AML-patient-derived primary cells”
What this piece can’t prove
  • Reporting is limited to an abstract; no methodological details (sample numbers, patient demographics, assay protocols, statistical analyses) are provided.
  • Experiments were ex vivo on primary cells; results do not establish in vivo safety, pharmacokinetics, or clinical efficacy.

1 further detail could not be confirmed from the summary.

Finally, the search trail

Method layer

NewsLink found the paper. Tessa takes you deeper.

NewsLink checks the story. Tessa is where you inspect the paper, authors, evidence, and research context.

Papers considered

The selected paper, plus nearby candidates.

PubMed, Crossref, Europe PMC · 35 candidate papers

Candidate

Kengo Kidokoro, author of, “Finerenone Improves Albuminuria via MR-TRPC Signaling in Diabetic Kidney Disease”

#HYPImpact · 2026 · Crossref

Source (opens in a new tab)
Candidate

Author response for "Leveraging Immunotherapy in Acute Myeloid Leukemia Treatment: Opportunities and Challenges"

2026 · Crossref

Source (opens in a new tab)

And 29 more candidates considered.