Economic Evaluation seminar - A multistate disease modeling approach to projecting population outcomes of Multicancer Detection Testing

Event details

Abstract: 

The rapid development in recent years of multicancer detection test (MCD) tests has led to questions about their potential for impacting population health. On the one hand, MCD tests offer the possibility to screen for many cancers that lack existing screening
programs. On the other hand, their clinical utility—including the balance of benefit and harms—is not yet known. Our team has taken a multistate modeling approach to project population benefits. We focus on a natural history model that divides the pre-clinical detectable period into early and late stages and can be used to project the reduction in late-stage disease under a screening program. To project mortality outcomes, we use a canonical model that links the reduction in late-stage disease to mortality improvements. This framework can be flexibly adapted to any solid cancer site and results combined across cancer sites to project the benefit of MCD screening.

In absence of screening trial data, we have calibrated our multistate model to age- and stage-specific incidence data and stage-specific survival data, in combination with key inputs about the sojourn time and sensitivity of the tests for early- and late-stage disease. Recently, we have extended our calibration approach to incorporate the pattern of screen and clinical diagnoses in clinical trials. Screening trial data provides a window in the pre-clinical period, providing a means to learn about pre-clinical sojourn time and sensitivity. We illustrate this calibration approach to the UKCTOCS trial of ovarian screening and discuss the potential to apply these methods to the upcoming NHS Galleri trial results. Limitations of the model will be discussed in addition to future model extensions.

If you are not a member of University of York staff and are interested in attending a seminar, please contact alfredo.palacios@york.ac.uk or shainur.premji@york.ac.uk so that you can be added to the mailing list.

Jane Lange, Staff Scientist, CEDAR, OHSU Knight Cancer Institute, School of Medicine

Web page