San Diego, CA – April 16, 2026 – SeekIn Inc., a global leader in affordable multi‑cancer early detection and precision oncology solutions, today announced to have four poster presentations at the AACR 2026 Annual Meeting, to be held on April 17–22 at the San Diego Convention Center in San Diego, California. The presentations span multi‑cancer early detection, patient‑specific treatment monitoring, AI‑empowered risk stratification, and health‑economic modeling of integrating multi‑cancer early detection tests into existing U.S. screening pathways.
“Our AACR 2026 posters collectively demonstrate a coherent vision: use widely accessible tools—routine lab tests, protein tumor markers, and cost‑effective blood‑based assays—to make early cancer detection and treatment monitoring scalable worldwide,” said Mao Mao, MD, PhD, Founder & CEO at SeekIn Inc. “From AI‑ empowered risk assessment to patient‑specific monitoring and rigorous cost‑effectiveness modeling, these studies show how our platform can be thoughtfully integrated into real‑world clinical workflows and national screening programs.”
OncoSeek 2.0: Advanced multi‑cancer blood test for early detection and diagnostic support
The first poster, titled “OncoSeek 2.0: an updated, cost-efficient blood-based assay with improved sensitivity for multi-cancer early detection” presents data on an upgraded version of SeekIn’s AI‑empowered multi‑cancer early detection (MCED) test. OncoSeek 2.0 incorporates three additional protein tumor markers to better capture prostate, lung, and squamous cell carcinomas on top of an established panel, while retaining the original machine‑learning framework.
Compared with OncoSeek 1.0, OncoSeek 2.0 achieved significantly higher AUC (0.860 to 0.917), with overall sensitivity improving from 65.3% to 75.9% at 90.1% specificity. The sensitivity gains were notable in prostate (+29.4%), head and neck (+26.6%), esophageal (+21.8%), cervical (+11.1%), and lung (+6.2%) cancers, addressing major causes of cancer death. Stage-stratified analyses further revealed that these improvements were concentrated in early-stage disease: stage I (+20.3%), stage II (+17.9%), and stage III (+15.8%), yielding an overall stage I–III sensitivity of 67.1%, while stage IV sensitivity remained consistently high for both versions.
This poster will be presented in the Clinical Research category, in the “Early Detection Biomarkers 1” session on April 19, 2026, 2:00–5:00 PM, at Poster Section 43, Board 6 (Poster 1096).
OncoGraph: Patient‑specific protein marker monitoring for recurrence and treatment response
The second poster, “OncoGraph: A patient‑specific blood test using protein markers to monitor cancer recurrence and treatment response,” introduces a low‑cost alternative to circulating tumor DNA–based minimal residual disease assays that leverages longitudinal trends in routinely measurable protein tumor markers (PTMs). OncoGraph first identifies abnormally elevated PTMs on a baseline blood sample—using OncoSeek as a front‑end screen—and then tracks those patient‑specific markers over time to signal recurrence risk or treatment response.
In liver cancer, patients who remained OncoGraph‑positive after surgery had substantially higher recurrence rates and shorter recurrence‑free survival than OncoGraph‑negative patients, indicating that the assay can provide early postoperative risk stratification. In advanced lung cancer and lymphoma cohorts, declines in patient‑specific PTMs after therapy initiation correlated closely with progression‑free survival and interim PET/CT response, respectively, suggesting that OncoGraph can offer concordant, less resource‑intensive monitoring in diverse tumor types. By relying on inexpensive PTMs and simple workflows, OncoGraph is designed to be deployable in low‑ and middle‑income countries where access to ctDNA‑based solutions is limited.
This poster will be featured in the Clinical Research category, in the “Biomarkers Predictive of Therapeutic Benefit 4” session on April 20, 2026, 2:00–5:00 PM, at Poster Section 41, Board 12 (Poster 3740).
AI‑empowered cancer risk assessment (CRA) using routine laboratory tests
The third poster, “AI‑empowered cancer risk assessment model based on routine laboratory tests for enriching individuals at high risk of cancer,” evaluates a random forest model that uses 56 features from standard blood and urine tests to estimate short‑term cancer risk. The model (OncoGate) was trained and validated on data from 7,672 individuals (1,245 cancer, 6,427 non-cancer) across two hospitals. An AI-based Personal Cancer Risk Score (PCRS), combining CRA and age, was further developed to efficiently enrich individuals at high cancer risk and then integrated with OncoSeek as a second‑step MCED test.
In the simulated cohort (overall incidence 0.57%), direct OncoSeek screening detected 2,450 cancers (PPV 5.6%). Age-based screening (≥50 years) tested 536,206 individuals and detected 2,327 cancers. PCRS risk–stratified screening tested 178,660 individuals and detected 2,402 cancers, with cancer incidence 2.69%. Compared with age-based screening, PCRS increased PPV from 7.6% to 15.9%, reduced false positives by 55.2%, and lowered cost per cancer detected from 18,436 USD to 5,950 USD, while maintaining overall detection.
This work will be presented in the Prevention, Early Detection, and Interception category, in the “Early Detection and Interception” session on April 21, 2026, 9:00 AM–12:00 PM, at Poster Section 37, Board 17 (Poster 5103).
Cost‑effectiveness of integrating MCED into U.S. screening programs
The fourth poster, “Cost‑effectiveness analysis of integrating multi‑cancer early detection into current cancer screenings in the United States,” uses a U.S. population‑level modeling framework to compare standard‑of‑care (SoC) screening alone with SoC augmented by different MCED strategies. The model incorporates 2021 SoC screening data for five major cancers (breast, cervical, colorectal, lung, and prostate), U.S. Census statistics, and real‑world cohort parameters for SeekIn’s two‑step MCED approach (OnceSeek Duet, OnceSeek followed by SeekInCare) as well as for a high‑priced comparator MCED test.
In a simulated cohort of 75.6 million screened individuals, SoC alone detected 197,773 cancer cases with positive predictive value (PPV) of 0.7% at a total cost of 43.2 billion USD, corresponding to 218,276 USD per detected case. Adding SeekIn’s OnceSeek Duet MCED test increased the number of detected cancers to 398,586—two times SoC alone—while improving PPV by 60-fold to 43.3% and reducing the cost per detected case to 135,429 USD, despite a higher total expenditure. A colorectal screening optimization scenario (fecal immunochemical test to replace primary colonoscopy, plus MCED) reduced SoC spending by 23.1 billion USD, fully offsetting the OnceSeek Duet costs, increasing detected cases to 391,022, and lowering cost per detected case to 78,899 USD (36.0% of SoC alone), without increasing the overall budget.
This poster will also appear in the Prevention, Early Detection, and Interception category, “Early Detection and Interception” session on April 21, 2026, 9:00 AM–12:00 PM, at Poster Section 37, Board 18 (Poster 5104).