Blood-Based Proteomic Profiling Identifies Osmr As a Novel Biomarker
Bibliographic record
Abstract
Background: Risk stratification in acute myeloid leukemia (AML) is critical to tailor timely induction therapy. The most widely utilized risk stratification approach is the European LeukemiaNet (ELN) that usually requires bone marrow biopsy and genomic testing. Inflammation is increasingly recognized as a critical factor in AML. Novel biomarkers from robust blood-based tests are needed to accurately and efficiently risk stratify patients with newly diagnosed AML. Methods: We evaluated the inflammatory secreted proteome through blood-based proteomic profiling of 251 soluble inflammatory proteins in 543 newly diagnosed AML patients to derive and validate a seven-protein prognostic score (Leukemia Inflammatory Risk Score, LIRS). Multivariable cox models with L1 regularization were used to test the independent prognostic ability. Individual proteins were evaluated as independently prognostic in multivariable cox models and model performance was assessed by cumulative concordance index (C-index) and time-dependent area under the curve (tdAUC). Findings were validated in internal and external cohorts, including a prospective cohort of newly diagnosed AML patients. Results: Serum from 362 newly diagnosed AML patients were collected prior to the administration of definitive induction therapy and profiled for 251 inflammatory proteins using NUcleic acid Linked Immuno-Sandwich Assay (NULISA), a proximity-ligation assay based on NGS or PCR allowing attomolar (10-18) detection level. The average detectability of inflammatory proteins was 97.3% across all 251 proteins. Accuracy of NULISA assay was validated by the correlation with known clinical variables and overlapping proteins measured in our clinical lab. To identify proteomic features with prognostic significance each protein was fitted into a univariate Kaplan-Meier analysis within the entire cohort. 148 proteins significantly associated with overall survival (OS) (adjusted p<0.05) were retained to build a regularized Cox model with LASSO regression to obtain the most predictive proteins. This led to the identification of 7 proteins strongly associated with OS: FGF23 (HR 2.11 95% CI: 1.60 - 2.79, p<0.001), GFAP (HR 1.91 95% CI: 1.44 - 2.52, p<0.001), IFNL1 (HR 1.66 95% CI: 1.27 - 2.21, p<0.001), MUC16 (HR 2.52 95% CI: 1.90 - 3.34, p<0.001), OSMR (HR 2.15 95% CI: 1.63 - 2.84, p<0.001), PDGFA (HR 0.67 95% CI: 0.51 - 0.88, p=0.0042), and VSNL1 (HR 0.58 95% CI: 0.44 - 0.76, p<0.001). The cohort was then randomly split into training (70%, n=245) and validation (30%, n=117) cohorts to define LIRS integrating these 7 proteins based on the coefficients of Cox regression model and validate the prognostic value of the score. LIRS was prognostic of OS in training and validation cohorts and remained prognostic when censoring for allogenic stem cell transplant (SCT) in first remission (CR1) in all and intensively treated patients. By multivariable adjustment, LIRS was independently prognostic after accounting for known prognostic factors in AML (HR 2.31 95% CI: 1.85 - 2.89, p<0.001), including age, ELN, creatinine etc. C-index and tdAUC suggested LIRS significantly outperformed ELN 2022 risk model. Individual proteins in LIRS were ranked, demonstrating OSMR, previously unrecognized in AML, as the most important prognostic protein. Increasing OSMR concentration led to consistent increase in hazard of death (HR 2.18 95% CI: 1.79 - 2.66, p<0.001) and remained significant when censoring for SCT in CR1 in all and intensively treated patients. Furthermore, OSMR as a single prognostic variable had a higher C-index than ELN 2022. OSMR was independently predictive of relevant clinical endpoints including early mortality and response rate to induction therapy. Additionally, OSMR is rapidly and easily detectable in the blood of newly diagnosed AML patients. LIRS and OSMR findings were validated in an external cohort of intensively treated patients (7+3, n=113) and prospectively in an internal cohort (n=68). Conclusions: Through high-throughput blood-based proteomic profiling, we identified a novel and strong prognostic signature LIRS with OSMR emerging as the best single biomarker that improve on current risk stratification guidelines for early and long-term risk of death in newly diagnosed AML patients (Patent Pending 63/573,150). This work adds important information for clinical translation to better inform AML patient risk.
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 imitationNot 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.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".