Impact of Pre-Transplant Comorbidities on the Rate of- and Mortality-Following Acute Graft-Versus-Host Disease (GVHD) After Allogeneic Hematopoietic Cell Transplantation (HCT)

Abstract 156

Pretransplant comorbidities, as captured by the HCT-comorbidity index (CI), predict an increased risk of non-relapse mortality (NRM) after allogeneic HCT. Whether the impact of comorbidities on NRM is due to increased incidences of HCT-related morbidities or their associated mortalities, or both is unknown. One of the main morbidities of allogeneic HCT is acute GVHD. Here, we investigated the impact of comorbidities on the rates of acute GVHD and post-GVHD mortality.

To this end, we analyzed data from 3033 consecutive patients (pts) treated with allogeneic HCT from HLA-matched related (55%) or unrelated (45%) donors at five institutions between 2000 and 2006. Median age was 45 (range 0.1–74.5) years. HCT-CI scores of 0, 1–2, 3, 4, and ≥5 were assigned to 32%, 31%, 18%, 8%, and 11% of pts, respectively. Diagnoses were myeloid (59%) and lymphoid (35%) malignancies or other non-malignant (4%) or malignant (2%) diseases. Conditioning regimens were high-dose (62%), reduced-intensity (15%), or nonmyeloablative (23%). A majority of pts received cyclosporine plus either mycophenolate mofetil (27%) or methotrexate (37%) for GVHD prophylaxis, while the remaining pts received either tacrolimus (18%)- or sirolimus-based (6%) or triple drug (12%) regimens. Stem cell source was marrow (20%) or peripheral blood mononuclear cells (80%). Proportional hazards models were used to estimate the hazards of acute GVHD and post-GVHD mortality and were adjusted for the covariates listed above. Increasing HCT-CI scores, modeled as a continuous linear variable, showed non-significant association with an increased risk of grades II-IV acute GVHD (p=.11). Alternatively, increasing HCT-CI scores were highly statistically significantly associated with an increased risk of grades III-IV acute GVHD (p <.0001). Compared to pts with a score of 0, those with scores of 1–4 and ≥5 were 1.49 and 2.16 times more likely to develop grades III-IV acute GVHD (p<.0001 for both) with respective rates of 13.2%, 18.4%, and 24.2%. Among pts who developed grades III-IV GVHD, the rates of mortality relative to a score of 0 were increased for scores of 1, 2, 3, and ≥4 (HRs=1.53, 1.49, 2.24, 2.98, respectively; p<.001 for each).

The HCT-CI was developed based on its association with NRM, not GVHD. Accordingly, we investigated the association of individual comorbidities and risk of acute GVHD. We studied each of the 17 comorbidities in the HCT-CI, plus 5 others (hyperlipidemia, gastro-esophageal reflux, hypothyroidism, hypertension, and coagulopathy). While none of these comorbidities was associated with HR >1.2 for grades II-IV acute GVHD, 11 comorbidities showed a HR >1.25 for grades III-IV acute GVHD (Table 1). With the same approach used for development of the HCT-CI, each of the 11 comorbidities was assigned a score of 1 in order to formulate a GVHD-specific CI. The resultant GVHD-CI scores were highly statistically significantly associated with increases in HRs and rates of grades III-IV acute GVHD and subsequent mortality (Table 2). The GVHD-CI predicted grades III-IV GVHD consistently among conditioning regimens.

Pre-transplant comorbidities predict grades III-IV acute GVHD and its subsequent mortality. A new concise GVHD-CI, comprising 11 relevant comorbidities, was developed to predict, at time of HCT, risks of grades III-IV acute GVHD and then predict, at time of diagnosis of GVHD, its subsequent mortality. Validation of the GVHC-CI is in progress. The set of comorbidities that comprise the GVHD-CI constitutes the basis for future research exploring the biology behind their association with acute GVHD, which would pave the way to individualized approach for prevention and treatment of acute GVHD.