The Joint Role of Comorbidity and Performance Status (PS) in Predicting Morbidity after Allogeneic Nonmyeloablative Hematopoietic Cell Transplantation (HCT).

An HCT-specific comorbidity index (HCT-CI) was developed to capture comorbidities among patients (pt) given HCT. Here, we asked whether HCT-CI could provide information independent from current measures, notably PS. PS before and toxicities after HCT were prospectively graded per Karnofsky and Common Toxicity Criteria, respectively, while comorbidities were retrospectively evaluated per HCT-CI. Spearman rank was used to evaluate correlations. This study included pts (n=107) diagnosed with hematologic malignancies and conditioned with nonmyeloablative regimen of 2 Gy total body irradiation alone (15%) or with 90 mg/kg of fludarabine (85%) before related (44%) or unrelated HCT (56%). There were no exclusion criteria for HCT-CI scores but pts with PS of <60% were excluded from HCT. Pretransplant parameters included age (median=56 years); prior regimens (median=4); interval between diagnosis and HCT (median=24.5 months); prior HCT (34%); low, standard, and high disease risks (10%, 63%, and 27%) as described (Kahl et al. Blood-2005).; PS scores (median=90% and range=60–100%); and HCT-CI scores (median=3 and range=0–7). After HCT, 44%, 9%, and 2% of pts developed grades II, III, and IV acute graft-versus-host disease (GVHD) and 24% and 15% developed overall grades III and IV toxicities of solid organs, respectively. There were no correlations between PS and age, prior regimens, interval before HCT, prior HCT, or disease risk. HCT-CI scores had weak correlations with prior regimens (r =0.16, p=0.09) and disease risk (p=0.03) and no correlations with age, interval before HCT, and prior HCT. Pts with high-risk disease and HCT-CI scores of 0–2 vs ≥3 had median prior regimens of 3 vs 5. There was inverse correlation between HCT-CI and PS scores (r = −0.32, p=<0.001). Both higher HCT-CI and PS scores predicted increased incidences of grades II-IV acute GVHD and grades III-IV HCT-related toxicities while only HCT-CI scores predicted increased risks of 1-year NRM and worse survival (Table 1). Table 2 illustrates a composite scale combining the HCT-CI and PS scores and its relationship with short-term morbidity. We conclude that the HCT-CI and PS are only weakly correlated and therefore both should be assessed at HCT. HCT-CI was a strong predictor of morbidity and mortality, while PS predicted morbidity only. The consolidation of both scales could provide a refined stratification of risk groups for HCT-related morbidity. Larger pt cohorts are required to better define such risk groups.