Abstract 1109: The complementary contributions of methylation and protein biomarker classes in a multi-cancer early detection (MCED) test

Abstract Background: Cancer is a complex disease driven by a multitude of molecular alterations. Multi-cancer early detection (MCED) tests rely on highly sensitive technologies to measure these diverse tumor-associated biomarker classes in blood. A multi-biomarker MCED testing approach may have the potential to detect more cancers than a single biomarker approach, as release of biomarkers into the circulation is driven by different mechanisms. We previously developed a multi-biomarker class (methylation and protein; MP V1) MCED classifier (Gainullin, medRxiv, 2025.08.24.25334244). Herein, we describe the contributions attributable to protein (P) and methylation (M) using a newly developed (MP V2) classifier. Methods: A prospectively collected case-control study (N=3,163; 729 cancers, 2,434 non-cancers) was used to develop a refined MP V2 classifier. MP V2 was trained to achieve robust early-stage sensitivity at a target specificity of ≥97.0%. M and P classifiers were trained individually; results were combined using a logical OR overarching classifier to generate the final call. MP V2 calls were deemed positive when either the M or P assay provided a positive call. We assessed the percentage of MP V2 positive calls due to P-only, M-only, and M+P (logical AND) positive calls. Results: Of the 378/590 positive MP V2 calls in the intended use (breast and prostate cancers excluded) cohort, M+P-positive calls were observed in 45.5%, P-only in 7.4%, and M-only in 47.1%. A P-only positive call was observed in 13.9%, 9.5%, 11.1%,6.5%, 5.7%, and 6.3% of stage I (N=36), II (N=63), stages I and II combined (N=99), III (N=123), IV (N=140), and unknown stage (N=16) MP V2 positive calls, respectively. An M-only positive call was observed in 63.9%, 65.1%, 64.6%, 52.0%, 28.6%, and 62.5% of stage I, II, stages I and II combined, III, IV, and unknown stage MP V2 positive calls, respectively. An M+P positive call was observed in 22.2%, 25.4%, 24.2%, 41.5%, 65.7%, and 31.1% of stage I, II, stages I and II combined, III, IV, and unknown stage MP V2 positive calls, respectively. Of the 64/2,434 (2.6%) false positive calls, none were positive for M+P; all false positives were a result of P-only or M-only positive calls. Conclusions: M and P biomarker classes provide independent contributions for MP V2 classifier positive calls, especially in early-stage (I and II) cases, demonstrating the potential value of adding proteins to methylation analysis in MCED testing. Prospective, real-world evaluation of the MP V2 classifier is currently underway. Citation Format: Vladimir G. Gainullin, Melissa Gray, Madhav Kumar, Stephen Luebker, Avinash Shanmugam, Kevin Cortes, Emily Chang, Philip J. Uren, Amin Mazloom, Jorge Garces, Tomasz M. Beer, Frank Diehl. The complementary contributions of methylation and protein biomarker classes in a multi-cancer early detection (MCED) test [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 1109.

Authors

Institutions

Publication Details

Journal
Cancer Research
Published
2026-04-03
DOI
https://doi.org/10.1158/1538-7445.am2026-1109
Primary Topic
Machine Learning in Bioinformatics
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Abstract 1109: The complementary contributions of methylation and protein biomarker classes in a multi-cancer early detection (MCED) test

Stephen A. Luebker, Amin R. Mazloom, Philip J. Uren, Jorge Garces et al.
Cancer Research
Machine Learning in Bioinformatics
article

Abstract 1109: The complementary contributions of methylation and protein biomarker classes in a multi-cancer early detection (MCED) test

Stephen A. Luebker, Amin R. Mazloom, Philip J. Uren, Jorge Garces, Tomasz M. Beer, Melissa A. Gray, Vladimir G. Gainullin, Frank Diehl, Avinash Shanmugam, Emily Chang, Kevin Cortes, Madhav Kumar
article en

Abstract

Abstract Background: Cancer is a complex disease driven by a multitude of molecular alterations. Multi-cancer early detection (MCED) tests rely on highly sensitive technologies to measure these diverse tumor-associated biomarker classes in blood. A multi-biomarker MCED testing approach may have the potential to detect more cancers than a single biomarker approach, as release of biomarkers into the circulation is driven by different mechanisms. We previously developed a multi-biomarker class (methylation and protein; MP V1) MCED classifier (Gainullin, medRxiv, 2025.08.24.25334244). Herein, we describe the contributions attributable to protein (P) and methylation (M) using a newly developed (MP V2) classifier. Methods: A prospectively collected case-control study (N=3,163; 729 cancers, 2,434 non-cancers) was used to develop a refined MP V2 classifier. MP V2 was trained to achieve robust early-stage sensitivity at a target specificity of ≥97.0%. M and P classifiers were trained individually; results were combined using a logical OR overarching classifier to generate the final call. MP V2 calls were deemed positive when either the M or P assay provided a positive call. We assessed the percentage of MP V2 positive calls due to P-only, M-only, and M+P (logical AND) positive calls. Results: Of the 378/590 positive MP V2 calls in the intended use (breast and prostate cancers excluded) cohort, M+P-positive calls were observed in 45.5%, P-only in 7.4%, and M-only in 47.1%. A P-only positive call was observed in 13.9%, 9.5%, 11.1%,6.5%, 5.7%, and 6.3% of stage I (N=36), II (N=63), stages I and II combined (N=99), III (N=123), IV (N=140), and unknown stage (N=16) MP V2 positive calls, respectively. An M-only positive call was observed in 63.9%, 65.1%, 64.6%, 52.0%, 28.6%, and 62.5% of stage I, II, stages I and II combined, III, IV, and unknown stage MP V2 positive calls, respectively. An M+P positive call was observed in 22.2%, 25.4%, 24.2%, 41.5%, 65.7%, and 31.1% of stage I, II, stages I and II combined, III, IV, and unknown stage MP V2 positive calls, respectively. Of the 64/2,434 (2.6%) false positive calls, none were positive for M+P; all false positives were a result of P-only or M-only positive calls. Conclusions: M and P biomarker classes provide independent contributions for MP V2 classifier positive calls, especially in early-stage (I and II) cases, demonstrating the potential value of adding proteins to methylation analysis in MCED testing. Prospective, real-world evaluation of the MP V2 classifier is currently underway. Citation Format: Vladimir G. Gainullin, Melissa Gray, Madhav Kumar, Stephen Luebker, Avinash Shanmugam, Kevin Cortes, Emily Chang, Philip J. Uren, Amin Mazloom, Jorge Garces, Tomasz M. Beer, Frank Diehl. The complementary contributions of methylation and protein biomarker classes in a multi-cancer early detection (MCED) test [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 1109.

Cancer ResearchVol. 86(7_Supplement)
Exact Sciences (United States) (US), IDEX Corporation (United States) (US)
Good health and well-being
Openalex Percentile: Top 30%
Machine Learning in Bioinformatics
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.