ZERO AT SUBMISSION: PICO FEASIBILITY FOR ADAGRASIB IN KRAS G12C-MUTATED NSCLC
Author(s)
Imen Reguei, PharmD1, Farah Moalla, BioE2, Lylia Chachoua, PharmD3, Mondher Toumi, MSc, PhD, MD4.
1Clever-Access, Tunis, Tunisia, 2Clever-access, Tunis, Tunisia, 3Clever-access, Paris, France, 4Aix-Marseille University, Marseille, France.
1Clever-Access, Tunis, Tunisia, 2Clever-access, Tunis, Tunisia, 3Clever-access, Paris, France, 4Aix-Marseille University, Marseille, France.
OBJECTIVES: Adagrasib received a conditional marketing authorisation for KRAS G12C-mutated non-small-cell lung cancer (NSCLC) on the strength of a single-arm trial. Using the 2024 JCA pilot exercise (MP03), this study asked how many of the scoped PICOs could be answered with the evidence available at submission, and how many were resolved by later data.
METHODS: We analysed the thirteen consolidated PICOs of MP03, mapped the single-arm registrational KRYSTAL-1 cohort against each, searched ClinicalTrials.gov and PubMed for comparators, and classified feasibility by a pre-specified algorithm anchored to the evidence existing at the conditional authorisation, distinguishing technically constructable comparisons from those methodologically acceptable under HTACG guidance.
RESULTS: The thirteen PICOs shared one intervention but spanned numerous overlapping populations — defined by prior therapy, histology, and PD-L1 status — against the most fragmented comparator landscape of the pilot exercises (sotorasib, docetaxel, platinum doublets, immune-checkpoint inhibitors, and several others). Counting comparator pairings and requested subgroups, the scope expanded to roughly 33 analyses, exceeding 40 at agent level. Because the pivotal cohort was single-arm, no PICO could be addressed by direct comparison; with no internal comparator to anchor a network, comparison would require unanchored methods, which HTACG guidance does not accept for relative-effect estimation. At submission, the result was zero feasible direct comparisons and zero methodologically acceptable indirect comparisons across all thirteen PICOs. A randomised confirmatory trial emerging after submission (adagrasib versus docetaxel) addressed only one PICO, with immature survival data; a later trial does not retroactively close the gap that existed at assessment.
CONCLUSIONS: For a single-arm-supported oncology product, an elaborate thirteen-PICO scope yielded no answerable comparison at submission, and the low certainty conclusion was predictable in advance — an argument for feasibility screening before a full dossier, and for a modified pathway that characterises uncertainty rather than documenting evidence absence.
METHODS: We analysed the thirteen consolidated PICOs of MP03, mapped the single-arm registrational KRYSTAL-1 cohort against each, searched ClinicalTrials.gov and PubMed for comparators, and classified feasibility by a pre-specified algorithm anchored to the evidence existing at the conditional authorisation, distinguishing technically constructable comparisons from those methodologically acceptable under HTACG guidance.
RESULTS: The thirteen PICOs shared one intervention but spanned numerous overlapping populations — defined by prior therapy, histology, and PD-L1 status — against the most fragmented comparator landscape of the pilot exercises (sotorasib, docetaxel, platinum doublets, immune-checkpoint inhibitors, and several others). Counting comparator pairings and requested subgroups, the scope expanded to roughly 33 analyses, exceeding 40 at agent level. Because the pivotal cohort was single-arm, no PICO could be addressed by direct comparison; with no internal comparator to anchor a network, comparison would require unanchored methods, which HTACG guidance does not accept for relative-effect estimation. At submission, the result was zero feasible direct comparisons and zero methodologically acceptable indirect comparisons across all thirteen PICOs. A randomised confirmatory trial emerging after submission (adagrasib versus docetaxel) addressed only one PICO, with immature survival data; a later trial does not retroactively close the gap that existed at assessment.
CONCLUSIONS: For a single-arm-supported oncology product, an elaborate thirteen-PICO scope yielded no answerable comparison at submission, and the low certainty conclusion was predictable in advance — an argument for feasibility screening before a full dossier, and for a modified pathway that characterises uncertainty rather than documenting evidence absence.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
MSR89
Topic
Methodological & Statistical Research
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, Oncology