Individualized prostate cancer testing : men's views on participation

Abstract

Prostate cancer (PCa) is the second most common cancer in men in the world and the fourth most common occurring cancer overall. However, because the harms from testing with prostate-specific antigen (PSA) are considered to outweigh the benefits, no governmental body has yet adopted a PSA-based screening program. Risk-based screening could potentially help reduce the proportion of men undergoing biopsy by identifying individuals at the highest risk of developing PCa, and thus reduce the harms of overdiagnosis and overtreatment. However, little is known about the psycho-social aspects surrounding risk-stratified PCa testing. This thesis, and the papers encompassed, aims to increase knowledge regarding men’s views on participation to individualized prostate cancer testing (PCT). Their interest in partaking in a risk-based PCT as well as its effect on their psycho-social health were investigated. A better understanding of their views, as well as predictors of participation and aspects of invitation would help inform development of population-based PCa screening programs to optimize attendance.

By using a cross-sectional survey in Paper I, the objective was to explore the general population’s interest in, and acceptability of, the prospect of risk-stratified cancer screening programs. A representative sample of 10.000 individuals (20-74 years of age) were invited to respond to a web-survey with questions developed by a panel of experts. Men were asked about PCa screening and women were asked about breast cancer screening. Out of our 2822 respondents (28%), a vast majority (94%) showed interest in wanting to know their cancer risk, with men presenting more certainty than women. A total of 87% agreed to the concept that if identified with a high risk, they would get screened more often. Only 27%, however, would agree to get screened less often if identified as having a low risk.

Paper II, Paper III and Paper IV studied actual participation in risk-based PCT. The PCa test was conducted within the frame of the STHLM3 trial, a large study for men 50-69 years of age in the region of Stockholm (Sweden). STHLM3 aimed at validating a risk-based PCT model in order to identify high-risk PCa. By participating in STHLM3, men were communicated their PCa risk (low, intermediate or high). The study sample in Paper II represented a sub-sample of 28.134 men invited to the pilot study of STHLM3. They were randomly allocated to different survey design factors in order to investigate optimization of participation rates. The study sample for Paper III and IV was also nested in STHLM3 and consisted of 10.000 men. They were invited to respond to a web- survey concerning worry, knowledge, health behavior and attitudes, as well as health related quality of life, three months before STHLM3, at invitation to STHLM3, and five months after participation in STHLM3.

Paper II and Paper III investigated predictors of participation to the risk-based PCa screening program. Paper II investigated survey and invitation design predictors (the use of a pre-notification, the length of the invitation letter, the length of the questionnaire and the use of a reminder or not). Paper III examined psycho-social predictors (worry, knowledge, health behavior and attitudes, and health-related quality of life) for participation in PCT. The participation rate in Paper II was 34%. The use of a pre-notification and a reminder increased participation to STHLM3. In Paper III, 1915 men responded to the questionnaire three months before invitation to STHLM3. When comparing decliners of STHLM3 (30%) with participants to STHLM3 (70%), participants presented more worry and an increased level of vulnerability, as well as a better general health than decliners.

Finally, almost 1000 men responded to the psycho-social questionnaire three months before STHLM3 as well as five months after STHLM3, enabling examination of the impact over time of participating to the risk-based PCT in STHLM3. Men assigned to a low or intermediate risk level reported that the levels of worry decreased over time, whereas men assigned to a high-risk level reported no increased level of worry. A low level of PCa knowledge was observed throughout Paper III and IV, calling for improved effort on that front before introducing PCa screening. Although participation rates could still be optimized, if implemented risk-stratified screening has the possibility to be accepted by the general public. Moreover, the study revealed no negative impact on the well-being of men participating in risk-based PCa testing.