CMV-antigenemia after allogeneic bone marrow transplantation: correlation of CMV-antigen positive cell numbers with transplant-related mortality.

One hundred and thirty-four consecutive patients undergoing HLA-identical BMT were prospectively followed on a weekly basis for the development of CMV antigenemia (CMVAg-emia). End-points of the study were (1) incidence, (2) risk factors, and (3) predictive effect on transplant-related mortality (TRM). Fifty-six patients developed CMVAg-emia between day 8-366 (median 40) with an overall actuarial risk of 43%. The median number of positive cells a diagnosis was 4 (range 1-48) the median maximum number was 6.5 (range 1-435). Positive cells are expressed as number/2.5 x 10(5) cells. In multivariate analysis, T cell depletion (TCD) (RR 2.9, P = 0.0009) and acute graft-versus-host disease (RR 2.1, P = 0.01) were the two risk factors predictive for CMVAg-emia. The risk of developing CMV-IP was significantly higher in patients with, as compared to patients without, CMVAg-emia (P = 0.0005) and occurred mostly in patients who received TCD marrow (P = 0.0009) despite treatment with gancyclovir or foscarnet at the time of CMVAg-emia. TRM was 24% in patients not developing CMVAg-emia; it was 21 and 47% in patients with "4" positive cells at diagnosis of CMV (P = 0.008), and 12 vs 54% for patients with "6" positive cells during infection (P = 0.0003). Both were predictive of TRM in multivariate analysis (P = 0.04 and P = 0.002). In conclusion, the risk of developing CMVAg-emia post-allo BMT is influenced by the marrow T cell content and by the occurrence of acute GVHD. High numbers of CMV antigen positive cells are associated with considerable transplant-related mortality, and may therefore identify patients eligible for early aggressive therapy.