Abstract

The hematopoietic cell transplantation-specific comorbidity index (HCT-CI) has been developed to identify patients at high risk of toxic mortality after an allogeneic stem cell transplant (alloSCT). Reduced intensity (RIC) and non-myeloablative regimens have decreased the non-relapse mortality (NRM) in heavily pre-treated and elderly patients. We performed a retrospective multicenter study to assess whether comorbidities, according to the HCT-CI, might influence the outcome of lymphoma and multiple myeloma patients undergoing RIC or non-myeloablative alloSCT. Two-hundred and three patients affected by non Hodgkin’s lymphoma (n=108), multiple myeloma (n=69) and Hodgkin’s lymphoma (n= 26) received an alloSCT from HLA matched sibling (n= 121) or unrelated (n=82) donors. Median age at transplant was 53 years (range, 17–69). The median number of previous chemotherapy was 3 (range, 0–7) and 68% of the patients received at least one autologous stem cell transplant (autoSCT). Twenty-five percent of the patients were transplanted in complete remission (CR), 50% in partial response (PR) and 25% in progressive disease (PD). RIC fludarabine-based regimens were used in 154 patients, whereas 49 patients received a non-myeloblative conditioning based on 2 Gy total-body irradiation+/− fludarabine. Variables included in multivariate analysis were age (<55 vs ≥55), HCT-CI (0 vs 1–2 vs ≥3), the Karnofsky Performance Status (PS) (>80% vs ≤80%), disease type (lymphoma vs myeloma), disease status before transplant (CR vs no-CR), the number of previous lines of therapy (≤2 vs >2), a previous autoSCT (0 vs ≥1), the donor type (sibling vs matched unrelated) and the conditioning regimen (non-myeloablative vs RIC). Patients with a HCT-CI of 0, 1–2 and ≥3 were 32%, 31% and 37%, respectively. The cumulative incidence of NRM was 5%, 16%, 20% at 1 year and 6%, 24% and 27% at 2 years, for patients with HCT-CI of 0, 1–2 and ≥3, respectively (p=0.04). The multivariate analysis for NRM showed that a high HCI-CI score (HR=1.60, p=0.03), as well as a low Karnofsky PS (HR=2.12, p=0.04) were correlated with a significantly worst outcome. Similarly, HCT-CI and the Karnofsky PS were able to predict overall survival (OS, HR=1.62, p<0.001 and HR=3.10, p<0.001, respectively) and unexpectedly, only HCT-CI retained significance in multivariate analysis for progression-free survival (PFS, HR=1.43, p=0.002). Univariate lymphoma subgroup analysis revealed that OS was better for patients with HCT-CI of 0 (p<0.001), with Karnofsky PS >80% (p<0.001), in CR at transplant (p=0.01) and receiving a RIC regimen (p=0.03). In myeloma patients, a previous autoSCT influenced OS (p<0.02) and there was a trend towards a significant correlation with HCT-CI of 0 and Karnofsky PS >80% (p=0.09 and p=0.07, respectively). When patients were analysed separately based on the conditioning regimen, OS was different for HCT-CI of 0, 1–2 and ≥3 either with RIC (p=0.001) or non-myeloablative regimens (p=0.02). Patients with HCT-CI 0, 1–2, and ≥3 had a similar NRM (p=0.19 for HCT-CI 0, p=0.87 for HCT-CI 1–2, p=0.33 for HCT-CI ≥3) and OS (p=0.94 for HCT-CI 0, p=0.76 for HCT-CI 1–2, p=0.18 for HCT-CI ≥3) when transplanted with non-myeloablative or reduced intensity conditioning. HCT-CI was inversely associated with Karnofsky PS (p<0.001, rho=−0.34) and the number of previous lines of therapy (p=0.002, rho=0.21), but not with age (p=0.38), time from diagnosis to transplantation (p=0.68) and pre-transplant disease status (p=0.73). Patients with a higher HCT-CI were not at higher risk of grade 2–4 acute GVHD (p=0.72) or chronic GVHD (p=0.77). These results demonstrated that HCT-CI may be a useful tool to predict NRM, OS and also PFS in lymphoma and myeloma patients undergoing RIC or non-myeloablative alloSCT.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author