Jane Lange
Staff Scientist
Research Interests
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
Articles (27)
Figure S3 shows posterior predictive checks for the primary analysis among HGSC cancers across all trial arms, comparing observed data with model predictions for clinical and screen-detected cancers.
Year:
2025
Figure S1 shows predictive model fits from approximate leave-one-out cross-validation for the full dataset and for cancer cases only, highlighting model selection steps and the final chosen model.
Year:
2025
Figure S7 shows posterior predictive checks for an analysis with episode-specific MMS sensitivities, comparing observed data with model predictions for clinical and screen-detected cancers across all trial arms.
Year:
2025
Supplementary Methods describe the model selection process using approximate leave-one-out cross-validation, posterior predictive checks comparing predicted and observed cancer incidence, and the estimation of empirical screening sensitivities from the fitted natural history model.
Year:
2025
Figure S6 shows the sensitivity of parameter estimates to model structure, displaying medians and credible intervals for onset rate, mean preclinical phase, and MMS/USS sensitivities across well-fitting models.
Year:
2025
Background: The UK Collaborative Trial of Ovarian Cancer Screening did not detect a reduction in ovarian cancer mortality with either multimodal screening (MMS) or transvaginal ultrasound screening (USS) compared with no screening. The trial data provide an invaluable resource to quantify the opportunity for interception in ovarian cancer. Methods: We used Bayesian inference to estimate ovarian cancer natural history based on individual screening and cancer diagnosis records from the UK Collaborative Trial of Ovarian Cancer Screening, a randomized controlled ovarian cancer screening trial conducted in England, Wales, and Northern Ireland. The trial included 202,638 women ages 50 to 74 years with no family history of ovarian cancer, randomized in a 1:1:2 ratio to annual MMS (serum CA125 interpreted using the risk of ovarian cancer algorithm), annual USS, or no screening. The current analysis included 199,499 women, with 674,806 screens and 2,025 cancer diagnoses. Results: Among high-grade serous cancers (HGSC), the estimated preclinical detectable phase was 1.7 years (95% credible interval, 1.3–2.2), compared with 7.8 years (95% credible interval, 5.7–10.6) for non-HGSCs. The preclinical detectable phase depended on screening modality: for HGSCs, it was longer in the MMS arm (2.2 years) compared with the USS arm (0.8 years), whereas for non-HGSCs, it was shorter in the MMS arm (2.7 years) compared with the USS arm (8.2 years). Conclusions: The interception opportunity for ovarian cancer strongly depends on histologic subtype and screening modality. Impact: Achieving a clinically significant benefit of ovarian cancer early detection will require prolonging the interception window through judicious combination of first- and second-line tests.
Year:
2025
Background: The sensitivity of a cancer screening biomarker to detect prevalent preclinical cancer drives screening benefit. Studies estimate sensitivity at different points in the biomarker development process. We examine how closely these estimates reflect the sensitivity to detect preclinical cancer (preclinical sensitivity). Methods: We posit that preclinical sensitivity is inversely proportional to the preclinical sojourn time. We simulate studies and estimates of sensitivity corresponding to the Early Detection Research Network’s Phases of Biomarker Development. Sensitivity estimates based on clinically diagnosed cases (phase II, clinical sensitivity), archived-sample studies (phase III, archived-sample sensitivity), and prospectively screened cohorts (phases IV and V, prospective empirical sensitivity) are defined and compared against the corresponding expected preclinical sensitivity. Results: Clinical sensitivity is generally optimistic. Archived-sample sensitivity is optimistic near clinical diagnosis but may be pessimistic at longer look-back intervals, with bias also dependent on test specificity. Prospective empirical sensitivity is optimistic when the sojourn time is long relative to the screening interval. Bias in prospective empirical sensitivity also depends on the frequency and accuracy of confirmation testing following a positive screening test. Conclusions: Sensitivity estimates from different phases of biomarker development should be distinguished and labeled accordingly to facilitate a realistic assessment of diagnostic performance and prediction of potential benefit. Impact: Our study highlights the need for clearer terminology to describe the sensitivity of cancer early detection biomarkers. We introduce new labels, explain biases in sensitivity estimates, and advocate for improved communication to enhance understanding of diagnostic test performance.
Year:
2025
There is a significant unmet need for clinical reflex tests that increase the specificity of prostate-specific antigen blood testing, the longstanding but imperfect tool for prostate cancer diagnosis. Towards this endpoint, we present the results from a discovery study that identifies new prostate-specific antigen reflex markers in a large-scale patient serum cohort using differentiating technologies for deep proteomic interrogation. We detect known prostate cancer blood markers as well as novel candidates. Through bioinformatic pathway enrichment and network analysis, we reveal associations of differentially abundant proteins with cytoskeletal, metabolic, and ribosomal activities, all of which have been previously associated with prostate cancer progression. Additionally, optimized machine learning classifier analysis reveals proteomic signatures capable of detecting the disease prior to biopsy, performing on par with an accepted clinical risk calculator benchmark.
Year:
2024
Background: Downstaging—reduction in late-stage incidence—has been proposed as an endpoint in randomized trials of multi-cancer early detection (MCED) tests. How downstaging depends on test performance and follow-up has been studied for some cancers but is understudied for cancers without existing screening and for MCED tests that include these cancer types. Methods: We develop a model for cancer natural history that can be fit to registry incidence patterns under minimal inputs and can be estimated for solid cancers without existing screening. Fitted models are combined to project downstaging in MCED trials given sensitivity for early- and late-stage cancers. We fit models for 12 cancers using incidence data from the Surveillance, Epidemiology, and End Results program and project downstaging in a simulated trial under variable preclinical latencies and test sensitivities. Results: A proof-of-principle lung cancer model approximated downstaging in the National Lung Screening Trial. Given published stage-specific sensitivities for 12 cancers, we projected downstaging ranging from 21% to 43% across plausible preclinical latencies in a hypothetical 3-screen MCED trial. Late-stage incidence reductions manifest soon after screening begins. Downstaging increases with longer early-stage latency or higher early-stage test sensitivity. Conclusions: Even short-term MCED trials could produce substantial downstaging given adequate early-stage test sensitivity. Impact: Modeling the natural histories of cancers without existing screening facilitates analysis of novel MCED products and trial designs. The framework informs expectations of MCED impact on disease stage at diagnosis and could serve as a building block for designing trials with late-stage incidence as the primary endpoint.
Year:
2024
Background: Downstaging—reduction in late-stage incidence—has been proposed as an endpoint in randomized trials of multi-cancer early detection (MCED) tests. How downstaging depends on test performance and follow-up has been studied for some cancers but is understudied for cancers without existing screening and for MCED tests that include these cancer types. Methods: We develop a model for cancer natural history that can be fit to registry incidence patterns under minimal inputs and can be estimated for solid cancers without existing screening. Fitted models are combined to project downstaging in MCED trials given sensitivity for early- and late-stage cancers. We fit models for 12 cancers using incidence data from the Surveillance, Epidemiology, and End Results program and project downstaging in a simulated trial under variable preclinical latencies and test sensitivities. Results: A proof-of-principle lung cancer model approximated downstaging in the National Lung Screening Trial. Given published stage-specific sensitivities for 12 cancers, we projected downstaging ranging from 21% to 43% across plausible preclinical latencies in a hypothetical 3-screen MCED trial. Late-stage incidence reductions manifest soon after screening begins. Downstaging increases with longer early-stage latency or higher early-stage test sensitivity. Conclusions: Even short-term MCED trials could produce substantial downstaging given adequate early-stage test sensitivity. Impact: Modeling the natural histories of cancers without existing screening facilitates analysis of novel MCED products and trial designs. The framework informs expectations of MCED impact on disease stage at diagnosis and could serve as a building block for designing trials with late-stage incidence as the primary endpoint.
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2024
Year:
2023
Year:
2023
Year:
2023
Year:
2023
Year:
2023
Collaborators (12)
Diana Miglioretti
Professor and Division Chief of Biostatistics
University of California, Davis
Nora Pashayan
Professor of the Epidemiology of Ageing
University of Cambridge
Ziding Feng
Professor
Fred Hutch Cancer Center
Spiros Denaxas
Professor of Computational Medicine
University College London
Sancy Leachman
Oregon Health & Science University
Yingye Zheng
Fred Hutchinson Cancer Research Center
Jason Xu
Duke University
Georgios Lyratzopoulos
Professor of Cancer Epidemiology
University College London
Usha Menon
Professor of Gynaecological Oncology
University College London
Matthew Barclay
Principal Research Fellow in Statistics
University College London
Juliet A. Usher-Smith
Associate Professor of General Practice
University of Cambridge
Aleksandra Gentry-Maharaj
University College London
Make your next move.
Prepare your application
Review drafts before you send them.