Consent, Concern, and Capacity: Factors Associated With Attitudes Toward Pharmacogenetic Testing Among Patients With Gastrointestinal Cancer.
Arravinth Anantharajah, Raza Chaudhry, Victoria A Wittner, Mari Angelica S Cayabyab, Glenda Hoffecker, Xingmei Wang, Laura J Conway, Penn Medicine Biobank, Ursina R Teitelbaum, Sony Tuteja
Abstract
Open AccessPharmacogenetics (PGx) applies knowledge of drug-gene interactions to maximize effectiveness and reduce toxicity of medications. PGx impacts the dosing of chemotherapy for gastrointestinal cancer, including fluoropyrimidines and irinotecan. However, pretreatment PGx testing is not routinely implemented due to barriers, including insufficient data surrounding counseling practices. This study assessed genetics knowledge and PGx attitudes in patients with gastrointestinal cancers. A survey was distributed to Penn Medicine Biobank and IMPACT-GI participants. Multivariable analysis was conducted to assess the influence of genetics knowledge and patient characteristics on attitudes toward PGx. 132 participants completed the survey, 69 (52%) female, 22 (17%) non-White, 50 (38%) colorectal and 40 (30%) pancreatic cancers. The mean genetics knowledge score was 2.67 ± 1.14 correct responses out of 4. Willingness to undergo pharmacogenetic testing was associated with higher genetics knowledge (β = 0.6, 95% CI [0.1, 1.0], p = 0.0132), being female (β = 1.1, 95% CI [0.0, 2.2], p = 0.0435), and higher self-reported health literacy (β = 2.2, 95% CI [0.9, 3.5], p < 0.001). Participants were less willing to undergo testing if they had health insurance concerns (β = -3.7, 95% CI [-5.8, -1.7], p < 0.001) or did not complete secondary education (β = -2.0, 95% CI [-3.5, -0.6], p = 0.006). Patients preferred providers to obtain consent and explain PGx's utility before testing. These findings provide initial patient-informed guidance for the implementation of pharmacogenetic testing in gastrointestinal cancer care and identify targets for patient education. As all survey respondents were research-engaged and from a single institution, the study population may not be widely generalizable.