“…5 Subsequently, a burgeoning body of research spanning pathophysiological, molecular, cellular and animal model studies has laid the initial groundwork, providing a biologically plausible basis for this conjectured relationship. [6][7][8][9][10][11][12] Nevertheless, the findings from clinical investigations, both in population-wide and clinical cohort studies, as well as the ensuing meta-analyses of such data, have presented divergent conclusions. [13][14][15] These discordant outcomes likely stem from several contributing factors: (i) the existence of crucial confounding variables, shared by individuals with cancer and OSA, such as chronological age, body mass index (BMI), smoking habits and other potentially overlooked sleep disorders, which demand comprehensive scrutiny 16 ; (ii) the retrospective nature of the majority of studies, which inherently imparts limitations regarding data quality and analysis [17][18][19][20][21][22][23] ; (iii) utilisation of national insurance health databases as primary data sources, which often lack key variables essential for comprehensive analysis 24 ; and (iv) inconsistent inclusion and assessment of OSA diagnosis and severity, often measured disparately as apnoea-hypopnoea index (AHI), oxygen desaturation indexes (ODI) 25 or mere nighttime oxygen saturation falling below 90% (TC90%) (notably, TC90% does not signify intermittent hypoxia but rather continuous nocturnal hypoxia, which may exist independently of OSA and could be contingent on underlying cardiovascular or respiratory conditions or obesity itself).…”