Rikke Andersen, Mette Syberg Jespersen, Mario Presti, Birgitte Bjørnhart, Anne-Cathrine Østby, Anne Kirstine Moeller Darras, Yago Garitaonaindía, Christina Halgaard Ruhlmann, Malene Støchkel Frank, Henrik Schmidt, Rasmus Blechingberg Friis, Louise Mahncke Guldbrandt, Jon Røikjær Henriksen, Andreas Bjerrum, Kira Schreiner Simonsen, Jesper Andreas Palshof, Troels Holz Borch, Jon Lykkegaard Andersen, Soeren Kjaer, Niels Viggo Jensen, Aziza Azimi, Ane Bundsbaek Iversen, Peter Meldgaard, Sara Grønbech Steen, Shawez Khan, A.A. Luczak, Stine Wahlstrøm, D. Zitnjak, Inge Marie Svane, Niels Fristrup, Charlotte Kristiansen, Gitte Fredberg Persson, Lars Bastholt, Mette Poehl, Eva Ellebaek, Marco Donia
PURPOSE: Among responders to PD-1/PD-L1 blockade, it remains unclear whether tumour type and traditional baseline patient and tumour characteristics provide additional prognostic information for progression risk beyond response depth. METHODS: In this nationwide, population-based cohort study, we identified adults with metastatic melanoma (MM), renal cell carcinoma (RCC), or non-small cell lung cancer (NSCLC) treated with PD-1/PD-L1 monotherapy or dual checkpoint blockade. Analyses were restricted to responders by investigator-assessed RECIST (complete or partial response) with outcomes administratively censored at 5 years. Prespecified baseline and on-treatment variables were evaluated for associations with progression-free survival (PFS) and overall survival (OS) using Kaplan-Meier and Cox models and assessed for consistency across tumour types. RESULTS: Among 2127 responders, PFS trajectories within response depth categories (complete response and partial response) were highly similar across MM, RCC, and NSCLC. Prespecified variables with prognostic value in the overall population, including performance status and treatment line, provided limited additional prognostic information once response depth was known. Depth of response (complete vs partial) was the strongest independent factor associated with progression risk within each tumour cohort. CONCLUSIONS: Among patients with MM, RCC, or NSCLC who achieved an objective response to PD-1/PD-L1 blockade, tumour type and traditional baseline prognostic factors provided limited additional stratification of progression risk once response depth was known. These findings suggest that, within the tumour types studied, response durability is primarily associated with response depth rather than other known clinical features, supporting prospective evaluation of response-guided follow-up strategies and interventions designed to deepen responses.