Psychosocial Evaluation in Allogeneic Hematopoietic Cell Transplantation Recipients (Allo HCT): Psychosocial Assessment of Candidates for Transplant (PACT) As a Tool to Identify High-Risk Patients and Its Association with Transplant Outcomes

Background: Allo HCT candidates undergo comprehensive evaluations including psychosocial assessment, although there are no validated tools to objectively assess psychosocial status prior to transplant. PACT, originally developed to address psychosocial risks in solid organ transplant recipients, has been evaluated by our group in allo HCT patients (Foster et al, BMT, 2009). It consists of 8 subscales which are scored from 1 to 5 (support stability, support availability, psychopathology, risk for psychopathology, health lifestyle, drug and alcohol use, compliance, and relevant knowledge). A final PACT score (range 0-4) provides an overall impression of a patient's suitability for transplantation.

Patients and Methods: This retrospective cohort study reviewed 404 adult allo HCT cases between 2003 and 2014 at Cleveland Clinic to identify predictors of adverse psychosocial status prior to allo HCT as determined by PACT. We then studied the association of PACT scores with allo HCT outcomes. Social workers generated a PACT score based on a comprehensive assessment of the patient. Median age of the study population was 50 (range 18-73) years, patients were 46% female, 11% non-White, 20% rural residents, and had mainly AML (41%) or MDS (18%). Majority of patients received HLA matched (84%) unrelated donor (55%) bone marrow grafts (55%) using myeloablative conditioning (78%).

Results: Final PACT rating was poor/borderline (score 0/1) in 5% (n= 21), acceptable (score 2) in 22% (87), good (score 3) in 44% (177), and excellent (score 4) in 29% (119) of recipients. In multivariable ordinal logistic regression, higher PACT score at pre-HCT assessment was associated with White race (OR 2.95, [95% CI 1.56-5.58], p<0.001), having a related donor (OR 1.61, [1.10-2.37], p=0.015), and a higher total QoL score (OR 1.22 per 10-point increase, [1.12-1.32], p<0.001). Final PACT rating was not associated with other variables evaluated including gender, place of residence (urban/rural), household income, diagnosis, disease risk, or time since diagnosis. Lower PACT score (0/1) was associated with poorer OS (57% versus 67% for PACT score 4), however this was not significant (P=0.11). A significant trend in 1-year non-relapse mortality (NRM) was observed (ranged from 33% to 16%, respectively, P=0.03); however, this association with NRM was not seen in multivariable analysis (HR 0.88, P=0.17). Patients with higher PACT ratings tended to spend more days outside the hospital alive in the first hundred days post-HCT in both univariate (P=0.07) and multivariable (P=0.09) analyses. We also analyzed individual PACT subscales with outcomes. None of the eight PACT subscales were associated with 1-year OS. However, "support availability" and "relevant knowledge" were associated with 1-year NRM in univariate analysis, although, only "relevant knowledge" remained significant on multivariable analysis (HR 0.81 per 1-point increase, [0.69-0.96], p=0.012) after adjusting for patient, disease, and transplant characteristics.

Conclusion: Non-White race, lack of a related donor, and lower QOL are associated with adverse psychosocial risk as measured by PACT prior to allo HCT. Low "relevant knowledge" and "final PACT" ratings were associated with NRM and suggest vulnerable populations. PACT can be effectively used as part of a comprehensive psychosocial assessment for identifying patients who may require additional psychological and social resources to help them navigate the transplant process.