e23148 Background: High-deductible health plans (HDHPs) are popular in the US due to their potential to curb rising healthcare costs and involve higher out-of-pocket costs for consumers which are associated with lower utilization of health services. Data on longitudinal trends for claims and cost responsibility for adult HDHP participants with cancer is scarce, which was the motivation for our study. Methods: We used multivariable Poisson regression models with the Huber/White/Sandwich linearized estimator of variance and predictive margins [reported as counts with 95% confidence intervals (CI)] to examine the effect of HDHP status on select stratified estimates of (a) the average number of claims per enrollee (claims), and (b) the average total participant responsibility per enrollee visit (cost in $; i.e., the sum of in-network and out-of-network values), for adult (age > / = 18 years) participants in the New Hampshire Limited Use Commercial Health Care Claims data, with breast and prostate cancer (BC/PC) diagnoses for the 10-year period 2010-19. Final models were fully adjusted for other plan types, year, age, and quintile of ecologic income, and our final dataset consisted of 945,281 claim records. Results: Overall, BC and PC HDHP participants had significantly fewer claims over the 10-year period [average counts: (BC: -242.2; -244.6, -239.8); (PC: -445.3; -450.3, -440.4)]. In addition, BC and PC HDHP participants had significantly lower participant responsibility over the 10-year period [average $: (BC: -85.4; -87.5, -83.2); (PC: -104.9; -115.5, -94.2)]. Overall trends for claims across the entire time period for HDHP vs. non-HDHP BC/PC participants were consistently lower, and flat trends were observed for BC from 2014-19 and for PC from 2012-19; however, there was steady increase in claims for both BC/PC non-HDHP participants across the same late periods. The trend for participant responsibility for BC HDHP participants was also lower compared with non-HDHP for all years except 2014; and after 2014 increases were seen for both BC HDHP and non-HDHP participants with the latter showing the steepest increases. The trend for participant responsibility for PC HDHP participants was lower compared with non-HDHP between 2010-17; but this relationship inverted for 2018-19 with HDHP participants showing the highest participant responsibility levels. Conclusions: Our findings show discordance in claims trends for HDHP vs. non-HDHP participants with BC/PC; durable trends of reduced claims are observed for HDHP participants, but for non-HDHP participants claim trends show steady increases in later periods.