HCT-CI Does Not Reliably Predict Outcome Following Allogeneic Stem Cell Transplant Due to Poor Correlation Between Clinical and Laboratory Assessment of Comorbidities

In recent years there has been considerable interest in the use of co-morbidity indices to risk stratify patients undergoing allogeneic stem cell transplantation (allo-SCT), although the optimal scoring system is unknown. A Haematopoietic cell Transplantation specific co-morbidity index (HCT-CI) has been developed to although the applicability of this index is not clearly established.

96 sequential patients (67 male, 29 female) transplanted at a single institution between 2008 and 2011 were included. Median age at transplant was 50 years (range 16–71). Indications for transplantation were acute leukaemia (n=49), MDS (n=13), myeloma (n=12), lymphoma (n=18), other (n=4). 33 patients received myeloablative conditioning that was either TBI based (n=22) or using chemotherapy alone (n=11). 63 patients received reduced intensity conditioning using fludarabine+TBI (n=8), fludarabine+melphalan+alemtuzumab (n=40). fludarabine+alkylator+ATG (n=13), other (n=1). 27 patients had undergone a prior transplant (autologous in 17, allogeneic in 6 and both in 4). Co-morbidities were assessed according to the HCT-CI by clinical evaluation together with measured assessment of organ function including estimation of left ventricular ejection fraction by echocardiogram or MUGA scan, pulmonary function tests (spirometry and gas transfer), creatinine clearance, together with liver (bilirubin, albumin, liver enzymes) and bone marrow function. Outcome was measured by estimation of non-relapse mortality and survival at 100 days and 1 year post transplant.

50 of 96 (52%) are currently alive at a median follow-up of 17 months (range 1–44). Mortality at day 100 and 1 year post transplant were 12.5% and 35% respectively. There was a trend to improved 1-year overall survival in patients with an HCT-CI score of zero compared to those with a score of >0 (log rank p=0.15) however the HCT-CI did not discriminate between patients with intermediate (1,2) and high risk (>2) scores. The number of patients with clinical co-morbidities was low; cardiac history in 5, diabetes in 2, obesity in 6, active infection at the time of conditioning in 4. The correlation between clinical and measured estimates of co-morbidity was poor particularly for respiratory function. One third of the cohort (32 of 96) had moderate to severe pulmonary dysfunction assessed by pulmonary function testing compared to 0 of 96 by clinical assessment. In multivariate analysis, there was no significant association between overall survival (at day 100 or 1 year post transplant) and any of the measured co-morbidity parameters used in this analysis; the number of co-morbidities as assessed by history or clinical examination was too small to allow any meaningful analysis.

In this single centre series the HCT-CI did not predict outcome of patients undergoing allo-SCT Patients referred for transplantation are highly selected and typically have little clinically apparent co-morbidity which hinders the utility of conventional risk scores such as the CCI for risk stratification. Laboratory tests are widely used for assessment of organ function although in this series, the results correlated poorly with outcome which questions their utility. Moreover, as applied within the HCT-CI they appear to overestimate co-morbidity (particularly pulmonary function) as compared to assessment by clinical history or examination. These results suggest that further work is required to identity a more reliable scoring system.

No relevant conflicts of interest to declare.