Introduction: Patients with hematologic malignancies are at increased risk of cognitive impairment following allogeneic blood or marrow transplantation (BMT) when compared with healthy controls. Cognitive deficits persist for at least 3-5 years in 40% of BMT survivors (Sharafeldin et al., J Clin Oncol, 2018); which negatively impacts self-management, relationships, quality of life, societal reintegration and overall survival. There is, thus, an unmet need to test interventions to improve cognitive function in BMT survivors. Here we report on the preliminary efficacy of a randomized wait-listed controlled trial at a single center, using home-based computerized cognitive training, in patients with hematologic malignancy treated with allogeneic BMT. We hypothesized that computerized cognitive training will result in significantly higher cognitive function scores in BMT recipients on the active intervention arm when compared with the waitlist arm. Methods: We prospectively enrolled a cohort of 53 adult (≥21y) allogeneic BMT recipients. Eligible patients were at least 3 months post-BMT at time of enrollment. The intervention used the Lumosity online cognitive training program for 12 weeks. The waitlist arm was offered the training program post-intervention. Cognitive function was measured at baseline and 12 weeks using standardized neuropsychological tests assessing executive function, attention, processing speed and memory cognitive domains. Study personnel were blinded to randomization assignment at time of baseline cognitive assessment. Participants' usage of the program was monitored to measure adherence to the intervention. Feasibility parameters included participation, completion, and attrition rates. We modeled the change in score between baseline and post-intervention by randomization group (active vs waitlist) as the main independent factor adjusted for baseline intervention test score as primary covariate. Models were further adjusted for cognitive reserve, age, sex, race, fatigue, sleep, depression, anxiety, and receipt of radiation. Results: The intervention had a 59% consent rate and a 79% randomization rate, 27 patients were randomized to active arm and 26 to waitlist arm. Mean age was 55y (SD=12.2y), 57.7% males, 84.6% non-Hispanic white; 57.7% with acute leukemia and 23% with myelodysplastic syndrome. Post-randomization attrition rate was 19.7% (17.6% on active arm; 20.5% on waitlist arm). Median adherence on the active arm for patients who completed 12 weeks of training was 87.5%; patients used the program for a median of 4.9 days/week and a median of 10/12 weeks. Compared to waitlist arm, we found improved scores for the active arm in global impairment deficit scores (0.08, SE=0.038, p-value=0.03) and in processing speed T-scores (3.43, SE=1.42, p-value=0.02). Conclusion: Our pilot study provides evidence for the feasibility and high adherence to a home-based cognitive training intervention in BMT survivors. Cognitive training as a compensatory intervention may be most efficacious in improving processing speed, however, additional interventions may be needed to further target other cognitive domains, e.g., executive function, attention, or memory domains.