Cognitive Outcomes Among Patients with Hematological Cancer with Access to Computerized Cognitive Training: Results from a Feasibility Study

Objective: Cancer-related cognitive impairment can affect patients after treatment of hematological cancers, but evidence of the effect of cognitive training interventions in this population is limited. The objective of this study was to explore cognitive outcomes among a sample of patients with hematological cancer who were given access to an online computerized cognitive training program after treatment completion.

Methods: Patients who had completed primary treatment for lymphoma, or autologous stem cell transplant for either lymphoma or multiple myeloma were recruited to test the feasibility of an 8-week computerized cognitive training intervention. All participants were assigned to receive the intervention, which was self-administered by participants at home and consisted of online brain training exercises designed to enhance information processing abilities. As part of the study, cognitive outcomes were assessed before (T1) and after (T2) the 8-week intervention, comprised of a standardized battery of neuropsychological tests, a self-report questionnaire of cognitive symptoms, and a functional assessment of medication management ability. Raw scores on neuropsychological tests were converted to T-scores based on demographically-adjusted norms (for age, sex, and education, where applicable), which were used for analysis. Composite T-scores representing performance in four cognitive domains (learning efficiency/memory, information processing speed/psychomotor efficiency, working memory, executive functioning) were computed at each time point. Wilcoxon's sum rank test was used to assess the differences in outcomes at times 1 and 2.

Results: 14 patients (12 lymphoma, 2 multiple myeloma) completed cognitive measures at T1 and T2. Participants were a median 48.5 years of age (range 24-70) and a median 53.5 days from treatment completion (range 28 - 490). At baseline, 7 participants met the criteria for overall cognitive impairment based on a Global Deficit Score (GDS) ≥0.5. After 8 weeks of access to the online computerized cognitive training program, there was a statistically significant improvement in the domains of information processing speed/psychomotor efficiency (MD = 4.18, std =4.55, p=0.0042) and working memory (MD = 2.09, std = 2.74, p=0.0127). Effect sizes (d) for changes in information speed/psychomotor efficiency and working memory were 0.92 and 0.76, respectively, suggesting large effects by Cohen's d classification. There was no sufficient evidence of a difference in learning efficiency/memory, executive functioning, cognitive symptoms or medication management ability.

Conclusion: Cognitive functioning in the domains of information processing speed/psychomotor efficiency and working memory were improved over the course of participation in an online self-administered computerized cognitive training program in this small sample of participants treated for hematological cancers. Post-hoc analyses to explore dose-response and factors associated with training time are forthcoming. Adequately powered studies with the inclusion of a control condition are required to determine treatment efficacy in this growing group of cancer survivors.