The Relationship between Depression, Anxiety, and Clinical Trial Perceptions Among Patients with Hematologic Cancer

Introduction. Despite the fact that thousands of cancer clinical trials (CCTs) are available today, engagement remains low, with only 2-7% of patients with cancer participating in CCTs. Research has shown that this may partially be due to fear-based perceptions around CCTs. Unfortunately, depression and anxiety, two psychological factors that are highly prevalent in the cancer space, are known to bias attention in ways that alter perceptions and are specifically known to amplify fear-based perceptions. Thus, the purpose of this exploratory study was to examine the relationship between depression, anxiety, and perceptions of clinical trials among patients with hematologic cancer.

Method: In this observational, cross-sectional study, 625 patients with hematologic cancer (46.4% multiple myeloma; 19.5% CLL; 11.4% non-Hodgkin lymphoma; 4.0% AML; 3.5% Hodgkin lymphoma; 3.2% CML; 1.6% ALL; 1.3% MPN; 7.4% other lymphoma; 1.8% other leukemia) completed the Cancer Support Community's online survey, the Cancer Experience Registry®. Participants provided sociodemographic and clinical history information, rated their level of agreement (0 = strongly disagree to 4 = strongly agree) with 8 statements related to beliefs about CCTs, and completed the Anxiety and Depression subscales (4 items each rated 1 = never to 5 = always) from the Patient-Reported Outcomes Measurement Information System (PROMIS-29v2.0). Responses to these 8 PROMIS items were averaged to compute a combined depression and anxiety score on a 5-point Likert scale. To understand the impact of depression and anxiety on perceptions of CCTs, 8 hierarchical regression models were examined; the dependent variable for each model was one of the CCT perception variables. Clinical history (cancer diagnosis, time since diagnosis, type of cancer care facility) and sociodemographic variables (age, gender identity, income, educational attainment, race, ethnicity, geographic area) were controlled for.

Results: The sample was 54.9% female, 86.7% Non-Hispanic White, 60.1 years old on average (SD=10.8) and had an average time since diagnosis of 5.3 years (SD=5.3; Median = 3.0 years; IQR = 6 years). 67.7% had a college degree, 20.5% had a gross annual household income of $100,000 or above, 41.4% received cancer treatment at an academic or comprehensive cancer center, and 45.6% lived in a suburban area. Participants' average anxiety and depression score was 1.91 (SD=.93).

Hierarchical regression analyses demonstrate that depression and anxiety had a significant effect on 7 of the 8 CCT perceptions assessed, when controlling for sociodemographic and cancer characteristics. Specifically, depression and anxiety were significant predictors of participants' perceptions that, "I would be unable to fulfill trial requirements due to logistical barriers" (ΔR ²=.019, b=.19, p=.003), "I don't trust the medical establishment and fear I will be used as a 'guinea pig'" (ΔR ²=.017, b=.17, p=.006), "I am uncomfortable with being randomly assigned" (ΔR ²=.016, b=.19, p=.01), "I fear receiving a placebo (for example, a sugar pill) in a clinical trial" (ΔR ²=.012, b=.18, p=.024), "I don't understand what clinical trials are" (ΔR ²=.011, b=.13, p=.021), "There are no clinical trials available in my community" (ΔR ²=.010, b=.14, p=.030), and "I fear side effects that might come with treatment on a clinical trial" (ΔR ²=.009, b=.13, p=.047). Thus, depression and anxiety accounted for significant amounts of variance in each of these clinical trial perceptions above and beyond the controls. Depression and anxiety did not have a significant impact on participants' perceptions that their health insurance would not cover a CCT (ΔR ²=.002, b=.05, p=.370).

Conclusion. Our findings demonstrate small but significant relationships between depression, anxiety, and perceptions of CCTs among patients with hematologic cancer. While common attempts to alter CCT perceptions often focus on information dissemination, the present study indicates that psychological factors may also need to be considered. While this study is an important first step in considering the relationship between mental health and perceptions of CCT, further longitudinal research is needed to better elucidate these findings. For example, differential analyses should explore if and how these relationships differ among patients with pre-existing clinically-significant levels of depression and anxiety.