MétaCan
Menu
← Back to cohort

Validation of an optimized tissue-agnostic genome-wide methylome enrichment assay to predict clinical outcomes in patients treated with pembrolizumab.

2025· article· en· W4410802882 on OpenAlexaff
Enrique Sanz Garcia, Eric Y. Stutheit-Zhao, Collin Melton, Junjun Zhang, Yongqi Zhong, Scott V. Bratman, Abigail Williams, Brian C. Allen, Jing Zhang, Daniel D. De Carvalho, Anne‐Renee Hartman, Zhihui Amy Liu, Albiruni Ryan Abdul Razak, Anna Spreafico, Philippe L. Bédard, Aaron R. Hansen, Stéphanie Lheureux, Pamela S. Ohashi, Lillian L. Siu

Bibliographic record

VenueJournal of Clinical Oncology · 2025
Typearticle
Languageen
FieldMedicine
TopicCancer Immunotherapy and Biomarkers
Canadian institutionsPrincess Margaret Cancer CentreUniversity of TorontoUniversity Health Network
Fundersnot available
KeywordsMedicinePembrolizumabDNA methylationGenomeOncologyComputational biologyInternal medicineBioinformaticsCancerGeneticsGeneBiologyImmunotherapyGene expression

Abstract

fetched live from OpenAlex

2545 Background: Recent work from the INSPIRE study (PMID38393391) suggests that kinetics of cell-free DNA (cfDNA) methylation profiles reflect immunotherapy treatment response in solid tumors. Here we provide validation data of a tissue-agnostic, genome-wide methylation enrichment assay based on cell free methylated DNA immunoprecipitation and high throughput sequencing (cfMeDIP-seq) designed for clinical use, to determine response to immunotherapy. Methods: This study utilizes samples and clinical data from the INSPIRE study, a single-institution investigator-initiated phase II study of pembrolizumab in multiple solid tumors given every 3 weeks (NCT02644369). A prior published analysis of cfMeDIP used TCGA to develop a classifier and demonstrated an association of response to immunotherapy. In contrast, in this analysis, a novel quantitative and highly specific measurement of ctDNA was estimated using a generative machine learning model trained on differentially methylated regions identified from a large cfMeDIP methylome atlas from individuals with and without cancer. In a blinded validation analysis, Firth’s logistic regressions were used to test differences in objective response (ORR) and clinical benefit rate (CBR) defined as complete or partial response or stable disease > / = 6 cycles between patients with a decrease in ctDNA from baseline to cycle 3 of treatment, and those with an increase in ctDNA. Sensitivity for no objective response, specificity for objective response, and positive and negative predictive values (PPV and NPV) were summarized. Cox regressions and log-rank tests were used to evaluate differences in progression-free survival (PFS) and overall survival (OS) between the two groups. Results: The analysis included 64 unique patients with a median follow up of 18.43 months (a total of 128 samples), including head & neck (n = 9), triple negative breast (n = 10), ovarian (n = 11), melanoma (n = 7), and other mixed solid tumor types (n = 27). A decrease in ctDNA was associated with significantly better objective response than an increase [odds ratio (OR) 33.89 (4.07, 44426.47), p = 0.0001], 58% sensitivity, 100% specificity, 100% PPV and 35% NPV. Significantly better CBR [OR 10.17 (2.74, 55.74), p = 0.0002] was also observed. A decrease in ctDNA was associated with significantly better PFS [hazard ratios (HR) 0.28 (0.15, 0.49) p < 0.0001] and OS [HR 0.42 (0.24, 0.76) p < 0.003]. Conclusions: A clinical tissue-agnostic, genome-wide methylome enrichment approach using cfMeDIP-seq accurately predicts clinical outcomes in patients treated with pembrolizumab in multiple advanced solid tumors. This test provides relative quantification of methylated ctDNA to predict response to immmunotherapy and does not require tumor tissue. This analysis highlights potential generalizability across tumor types in response monitoring. Clinical trial information: NCT02644369 .

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.002
metaresearch head score (Gemma)0.004
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Bench or experimental · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.002
Threshold uncertainty score0.011

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0020.004
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0010.000
Science and technology studies0.0000.000
Scholarly communication0.0010.000
Open science0.0000.000
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0010.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.051
GPT teacher head0.439
Teacher spread0.388 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designBench or experimental
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueJournal of Clinical Oncology→Same topicCancer Immunotherapy and Biomarkers→French-language works237,207→