{
  "status": "Descriptive external benchmarks only; no parameters fitted or validated.",
  "reviewed": "2026-09-18",
  "model": {
    "screen_detected_multiplier_range": [
      1.882752476069576,
      3.568418421323743
    ],
    "incremental_cancers_per_1000_year_range": [
      1.3194546367945335,
      6.406382947170057
    ],
    "population": "Separate age/sex groups 50-74, full adherence, steady-state annual scenarios"
  },
  "pathfinder_2": {
    "source": "https://grail.com/wp-content/uploads/2026/05/Pathfinder2_FactSheet_FINAL.pdf",
    "additional_source": "https://www.asco.org/abstracts-presentations/260912",
    "design": "Single-arm prospective study, initial Galleri test and 12-month follow-up, age 50+",
    "performance_analyzable_n": 32007,
    "usual_ab_detected": 31,
    "galleri_detected": 173,
    "galleri_new_primary": 151,
    "galleri_recurrent": 22,
    "combined_screen_detected_ab": 204,
    "route_based_multiplier": 6.580645161290323,
    "galleri_detected_per_1000_participants": 5.405067641453432,
    "new_primary_galleri_detected_per_1000_participants": 4.717717999187678,
    "false_positive_results": 114,
    "false_positive_results_per_1000_participants": 3.561720873558909,
    "limitation": "Detection-route accounting is not randomized incremental yield. Includes recurrent cancers; uptake and screening phase differ from the model."
  },
  "nhs_galleri": {
    "source": "https://grail.com/wp-content/uploads/2026/05/NHS-GalleriASCO_Factsheet_FINAL.pdf",
    "design": "Randomized UK trial, three annual rounds, ages 50-77",
    "control_screen_detected": 290,
    "intervention_usual_screen_detected": 236,
    "intervention_galleri_detected": 937,
    "intervention_total_screen_detected": 1173,
    "screen_detected_count_ratio": 4.044827586206897,
    "screen_detected_count_difference": 883,
    "reported_false_positive_rate": 0.0045,
    "limitation": "Count difference across trial arms, not annual age-specific yield. UK eligibility, uptake, baseline screening and repeat rounds differ."
  },
  "interpretation": "Trials support additional screen detection with low marginal false-positive rates. Their larger multipliers do not establish that the model is conservative or validate its age estimates. Age/sex-specific trial incidence, attendance, prior screening and detection-route data are needed for matched calibration."
}