Monitoring Immune Function in the Allogeneic Transplant Recipient: Predicting Graft vs Host Disease (GvHD) and Infection.

Background: GvHD, intensity of immunosuppression (ISP) and infection are major causes of toxicity post allogeneic stem cell transplantation (allo-SCT). Yet, no method reliably predicts these complications. We explored the use of a novel FDA-approved immune function assay to determine whether an immunologic response to changes in immunosuppressive therapy for GvHD could predict risk of GvHD and infection and allows tailoring ISP and infection prophylaxis. The ImmuKnow is an assay of global T-cell function that measures increased ATP production after stimulating CD4+ T-helper cell and has been shown useful for managing ISP therapies in solid-organ transplantation.¹ We hypothesized that ImmuKnow could assist in predicting risk of GvHD and infection.

Patients and Methods: Between March 2004 and March 2006, 24 adult allo-SCT patients (pts) (mean age 48 years; range 23–70; 50% females) underwent blood sampling for determining ImmuKnow at various times post-HSCT (267 assays, mean of 11/pt). Results were correlated with clinical course (infection, GvHD or event-free). Underlying diseases included acute myelogenous leukemia (n=9), chronic myelogenous leukemia (n=4), myeloma (n=3) and others (n=8).

Results: during the testing period, GvHD was documented in 10 pts [(acute, n=3; chronic, n=7] and 20 infections occurred [bacteremia (n=9); CMV viremia (n=4); others (n=7)]. The average immune function (ng/mL ATP) of GvHD pts was significantly higher than that of event-free pts (385 +/− 102 vs. 189 ng/mL +/−78 respectively), p<0.001. Pts with infections (13 pts, 71 events) had a mean 86 ng/mL ATP (+/−87), also statistically different (p<0.001) from event-free. ROC analysis showed that the positive predictive value (PPV) for GvHD with ImmuKnow > 279 ng/mL was 74 with a sensitivity of 90% (95% CI, 76.9–97.2) and specificity of 83% (95% CI 73 - 91) and an odds ratio (OR) for GVHD of 42 (95% CI, 13 - 136, p<0.001). For infection, ImmuKnow values < 94 ng/mL were 70% sensitive (95% CI, 59 - 80) and 87% specific (95% CI, 77 - 94) with a PPV of 85 and OR of 17 (95% CI, 8–39, p<0.001).

Conclusions: these preliminary data suggest that the ImmuKnow assay is an objective marker of risk for infection and GvHD post-HSCT. Tailoring intensity and duration of ISP to reduce risks of GvHD and infection in an individual patient may be possible and may improve outcomes. Validation of these preliminary data is ongoing.