Background: Patient preference studies capture the value patients place on treatments, outcomes, and care delivery. Patient preference information (PPI) is critical to inform patient-focused drug development, care delivery, and regulatory decision-making such as FDA's benefit-risk assessment of novel therapies. There are currently no published systematic reviews consolidating PPI for blood cancers. The objective of this study was to identify and appraise existing published PPI from patients with blood cancers and generate a pooled estimate of the relative importance of treatment outcomes.

Methods: We performed a systematic review (PROSPERO 2022 CRD42022364647) and reported findings in accordance with PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines. Articles were included from PubMed and Embase. Included articles reported primary data on the preferences of patients with blood cancers (or patient proxy). All abstracts were independently reviewed by two authors. Group discussion resolved conflicts. Quality was assessed according to the PREFS checklist and risk of bias was assessed. Data extraction covered study type, method, and PPI. Attributes included were catalogued and classified by domain. PPI from discrete choice experiments (DCEs) that included attributes in the domains of survival/treatment response, quality of life/side effects, and care delivery/cost were included in meta-analysis. Relative attribute importance (RAI), the proportion of decision influence attributed to each domain, was quantified and pooled to compare data across studies.

Results: A total of 3,788 abstracts were screened; 386 underwent full-text review, and 119 were included in data extraction. Most (83%) were published in the last 10 years. In total, 26,353 cancer patients and 776 patient proxies were represented across studies. Studies collected PPI across disease types: multiple myeloma (n=50, 42%), lymphoma (n=32, 27%), acute myeloid leukemia (n=26, 22%), chronic lymphoid leukemia (n=16, 13%), chronic myeloid leukemia (n=12, 10%), and acute lymphoblastic leukemia (n=10, 8%). Preference elicitation methods included study-specific surveys (n=51, 43%), discrete choice experiments (n=39, 33%), and qualitative interviews (n=24, 20%). Studies reported PPI for treatment outcomes (n=83, 70%), route of administration (n=23, 19%), and communication (n=19, 16%). Studies focused on informing clinical decisions (n=94, 79%), followed by policy (n=23, 19%), and regulatory decision-makers (n=11, 9%). Funding for the studies came from pharmaceutical companies (n=47, 39%) government grants (n=23, 19%), patient advocacy organizations (n=16, 13%), and cancer centers (n=10, 8%). Most studies were of high (n=94, 79%) or moderate (n=19, 16%) quality with some studies having apparent (n=22, 18%) or likely (n=24, 20%) conflicts of interest. Preference heterogeneity was reported across studies.

Meta-Analysis: The weighted RAI from combined data from 28 DCEs demonstrated that Surivival/Treatment response and Quality of life/Side effects were equally important (0.40, 95% confidence interval [CI] 0.34 – 0.47; 0.39, 95% CI 0.33 – 0.46). These domains were twice as important compared to Care delivery/Cost (0.20, 95% CI 0.16 – 0.25).

Conclusions: A growing evidence base over the past decade has resulted in a substantial amount of PPI on blood cancers. Data from this meta-analysis demonstrate that achieving treatment efficacy had similar importance to patients with blood cancers as avoiding side effects/maintaining quality of life; care delivery/cost was half as important as either of these outcomes. PPI from studies included in this systematic review can be used to understand benefits and risks from the patients' perspective, clarify how patients consider tradeoffs between important outcomes, and inform the design of future patient preference studies. The substantial preference heterogeneity seen in all studies suggests that most treatment decisions for patients with blood cancers are preference-sensitive based on the FDA definition. This suggests that including PPI in regulatory submissions may be beneficial to trial sponsors to aid the FDA in benefit-risk assessment. Although many studies highlighted the need to incorporate PPI into care delivery, further work is needed to develop strategies to achieve this aim.

This content is only available as a PDF.
2025
Sign in via your Institution