Presentation (CTI)

OBJECTIVES: Canada’s two oncology health technology assessment bodies, Canada’s Drug Agency (CDA-AMC) and National Institute of Excellence in Health and Social Services (INESSS), share similar guidance but have distinct mandates. Manufacturers often file in parallel, and there is growing interest in understanding how recommendations and price expectations diverge to guide evidence generation strategies.
METHODS: We reviewed paired oncology submissions from January 2024-December 2025, excluding single-agency filings and ongoing reviews without public summaries. We abstracted date, indication, recommendation, price conditions, model re-analysis/modifications, and recommended price reduction. French INESSS text was translated to English; reductions were summarized with five-number statistics. Paired differences were computed as CDA-AMC and INESSS.
RESULTS: Among 57 matched submissions, CDA-AMC and INESSS refused two and eight, respectively; 10 were cost-minimization. Recommended price reductions were summarized as (min, Q1, median, Q3, max): CDA 2.5%, 45.5%, 76.0%, 83.5%, 89.0% (n=35); INESSS 0%, 40.0%, 66.0%, 86.5%, 100% (n=39); and paired differences (CDA − INESSS) -61.7%, -25.0%, -7.0%, 5.0%, 77.0% (n=29), where more negative values indicate INESSS requested larger cuts. The time trend across 29 paired submissions, each reporting reductions over 24 months, increased slightly from -0.45 to -0.25; linear fit was essentially flat (slope 2.2×10⁻⁵/day, R²≈0.0006), indicating no meaningful time trend.
CONCLUSIONS: Observed differences likely reflect agency‐led re-analysis/modifications (e.g., population/line restrictions, survival modelling, utilities, cost inputs) rather than simple differences in “price aggressiveness.” CDA-AMC and INESSS may assess different model bases (manufacturer vs re-analyzed), therefore, reductions are not strictly like-for-like. Recent months show more INESSS refusals, suggesting increased stringency/aggressiveness, although this pattern requires confirmation with larger samples. Refusals typically omit price-reduction figures, constraining paired comparisons and potentially biasing estimates toward approved or conditional listings. Priorities include grading re-analysis severity/targets and linking modifications to incremental cost-effectiveness ratio shifts and reduction magnitudes. Given n=29 for paired differences and evolving report formats, findings warrant cautious interpretation.