Biomarker Prediction

ScientificConcept

The emerging capability to predict whether blood biomarkers (such as cortisol) are out of range purely by analyzing continuous non-invasive physiological signals from the wrist.


First Mentioned

11/1/2026, 5:22:55 AM

Last Updated

11/1/2026, 5:29:58 AM

Research Retrieved

11/1/2026, 5:29:58 AM

Summary

Biomarker prediction refers to the process of forecasting future health indicators, potentially without invasive procedures. Companies like Whoop are pioneering this field by training foundational models on extensive physiological data, uploaded medical records, and blood tests to predict when biomarkers such as cortisol might be out of range. This technology aims to provide insights into health status by analyzing continuous monitoring data from wearable devices, which track metrics like heart rate variability, respiratory rate, and skin temperature. By correlating this sensor data with user-logged habits and other health information, these systems can offer personalized guidance and potentially detect serious health issues early. For example, Whoop's system has alerted members to potential health problems like chest and lung inflammation, COVID-19, stage four Hodgkin's lymphoma, and testicular cancer by identifying deviations from their normal physiological baselines. The ultimate goal is to move towards more proactive and preventative healthcare by predicting health outcomes and biomarker levels before they become critical.

Research Data
Extracted Attributes
    Prediction

    A prediction (from Latin prae- 'before' and dictum 'something said') or forecast is a statement about a future event or about future data. Predictions are often, but not always, based upon experience or knowledge of forecasters. There is no universal agreement about the exact difference between "prediction" and "estimation"; different authors and disciplines ascribe different connotations. Future events are necessarily uncertain, so guaranteed accurate information about the future is impossible. Prediction can be useful to assist in making plans about possible developments.

    Web Search Results
    • Predictive Biomarkers: Matching the Right Treatment to the Right Patient - Qualisure Diagnostics

      While prognostic biomarkers tell us how a cancer is likely to behave, predictive biomarkers tell us something different and arguably more actionable. They tell us whether a specific treatment is likely to work for a specific patient. That distinction matters enormously in clinical practice, and it is what makes predictive biomarkers the foundation of truly personalized cancer care. ## What Makes a Biomarker Predictive? A predictive biomarker is a biological feature that can identify patients who are more likely to respond to a particular therapy. It does not simply describe the cancer. It connects a measurable characteristic of the tumor to an expected treatment outcome, essentially acting as a guide for which treatment path makes the most sense for that individual. [...] ## From Discovery to Clinical Use: The Importance of Validation Discovering a potential predictive biomarker is only the beginning. Before it can be used to guide real treatment decisions, it needs to be rigorously validated, meaning researchers need to confirm that it works reliably and consistently across many different patients and settings. This typically involves large clinical trials specifically designed to test the biomarker’s performance, not just the treatment itself. Researchers need to show that patients who test positive for the biomarker respond better to the treatment, and that patients who test negative do not benefit in the same way. That comparison is what makes a biomarker clinically useful. [...] The test used to detect the biomarker also needs to be dependable. It should produce consistent results whether it is run in a hospital laboratory in Calgary or a research centre in Singapore. If the test performs differently depending on where or how it is used, the biomarker loses its practical value. In many cases, a predictive biomarker and its companion test are reviewed and approved together by regulatory agencies, such as Health Canada or the US Food and Drug Administration, before they become part of routine clinical practice. ## Examples in Practice Several predictive biomarkers are now well established in oncology and used routinely to guide treatment decisions.

    • MarkerPredict: predicting clinically relevant predictive biomarkers with machine learning | npj Systems Biology and Applications

      achieving a 0.7–0.96 LOOCV accuracy. We defined a Biomarker Probability Score (BPS) as a normalised summative rank of the models. The scores identified 2084 potential predictive biomarkers to targeted cancer therapeutics, 426 was classified as a biomarker by all 4 calculations. We detailed the biomarker potential of LCK and ERK1. This study encourages further validation of the high-ranked predictive biomarkers. The development of the MarkerPredict tool (which is available on GitHub) for predictive biomarker identification may have a significant impact on clinical decision-making in oncology. [...] Two examples illustrate how the MarkerPredict framework identifies both established and novel predictive biomarker candidates. The first is LCK, a Src-family kinase predicted as a biomarker for EGFR inhibitors. LCK is involved in cytoskeletal organisation and integrin signalling and has been shown to support compensatory EGFR signalling in cholangiocarcinoma43.") Its predictive role has also been supported in clinical cohorts of lapatinib-treated breast cancer patients42."), further validating the model’s performance. The second example is ERK1, a novel prediction as a biomarker for BCL2 inhibitors such as venetoclax. While not currently used as a clinical biomarker, ERK1 regulates multiple pro-apoptotic BCL2 family members and mediates resistance to BCL2 inhibition via stabilization of [...] With the MarkerPredict method, we established a high-accuracy binary classification algorithm for the prediction of new predictive biomarker–therapy target pairs based on network topological and protein annotation data. This approach is novel. However, many algorithms were developed to classify and analyse biomedical data49."). Here we show a few examples of the Random Forest and XGBoost algorithms used in cancer research and biomarker identification.

    • IDENTIFYING BIOMARKERS FOR TREATMENT RESPONSE PREDICTION | Netherlands Cancer Institute

      Precision oncology is a targeted approach to diagnosing and treating cancer based on an individual’s specific profile. In personalised cancer therapy, biomarkers (measurable and quantifiable biological molecules that indicate a specific biological state or condition) are used to predict the course of the disease (prognostic biomarkers) and which patients will have the highest probability (predictive biomarkers) of responding or having adverse side effects to particular therapies. Our team focuses on developing Computational Pathology approaches to find and validate biomarkers to be implemented in daily clinical practice to empower precision oncology. [...] Home Hugo Horlings Research Research Groups Hugo Horlings Our research # IDENTIFYING BIOMARKERS FOR TREATMENT RESPONSE PREDICTION Cancer is a complex disease and although overall mortality rates have decreased over the past decade, the total number of new diagnoses per year continues to rise. This highlights the urgent need for improved methods to identify individuals who would benefit from effective cancer treatments while minimizing treatment-related side effects. Moreover, with advancements in surgical techniques, radiotherapy, and various systemic therapies such as chemotherapy, targeted therapy, and immunotherapy, there is a growing opportunity and necessity for personalized cancer treatment based on biomarker-driven approaches, also known as precision oncology.

    • Biomarker Discovery and Validation: Statistical Considerations

      A predictive biomarker needs to be identified in secondary analyses using data from a randomized clinical trial, through an interaction test between the treatment and the biomarker in a statistical model. Secondary analyses refer to subsequent correlative studies which may or may not be pre-defined as a protocol objective. An example of predictive biomarker identification is the IPASS study 16. The IPASS study enrolled patients with advanced pulmonary adenocarcinoma who were nonsmokers or former light smokers and randomly assigned patients to receive gefitinib or carboplatin plus paclitaxel (CP). Patients’ EGFR mutation status was not known at the time of enrollment and was determined retrospectively. The interaction between treatment and EGFR mutation was highly statistically significant [...] A biological marker (biomarker) is “a defined characteristic that is measured as an indicator of normal biological processes, pathogenic processes, or biological responses to an exposure or intervention, including therapeutic interventions” 1. Biomarkers have various applications such as risk estimation, disease screening and detection, diagnosis, estimation of prognosis, prediction of benefit from therapy, and disease monitoring (Figure 1). In oncology, biomarker candidates often consist of biological molecules found in cancer cells. The most common biomarkers are cancer-associated proteins, gene mutations, deletions, rearrangements and extra copy numbers of genes. These molecules are sometimes secreted into the circulation and so may be detected by blood-based assay, while others are

    • Validation of Biomarker-based risk prediction models

      There is vast statistical literature on how to build predictive models (e.g. 13, 14). Multivariable models allowing simultaneous association of biomarkers and predictors with clinical outcome, such as logistic regression for presence/absence of disease or Cox regression with survival, are common building blocks of biomarker-based risk prediction tools. More algorithmic statistical models, such as support vector machines or neural networks, can also be used.