Abstract
Background: Up to 70% of allogeneic stem cell transplant (alloSCT) recipients who are cytomegalovirus (CMV) seropositive experience CMV reactivation and up to a third are complicated by end-organ disease. Letermovir, an antiviral agent targeting CMV, is FDA-approved for prophylaxis of CMV reactivation in CMV seropositive recipients of alloSCT. 1-4 In CMV seropositive recipients, letermovir use up to week 14 after alloSCT has demonstrated a significantly lower incidence of clinically significant CMV infection (csCMVi) and an overall mortality benefit as far as 24 weeks after transplant. 5-8 However, data on csCMVi beyond 24 weeks is lacking. Additionally, more information is needed on letermovir use in patients who had prior alloSCT, are CMV seronegative, have a detectable CMV viral load but without csCMVi, and/or are high risk*.
Methods: This is a single-center, retrospective, comparative cohort analysis of 524 patients who received alloSCT at Barnes-Jewish Hospital from January 2016 to June 2019. Of those, 191 patients were excluded because both the recipient and donor were seronegative for CMV. Patient information was obtained from the electronic medical record systems after IRB approval of the protocol. Gray's sub-distribution methods (while account for death as competing risk) were used to calculate the incidence of csCMVi and to assess the effect of letermovir on csCMVi. Univariate and multivariate Cox proportional hazards models were fitted for the effect of letermovir on overall survival (OS), and time-dependent Cox model was used to determine the effect of csCMVi on OS.
Results: Out of 333 patients, 149 received letermovir. The median follow-up period was 13.38 months (0.033 to 63.8 months). A univariate analysis demonstrated that csCMVi was associated with worse OS (HR 2.173, 95% CI 1.602-2.948). Among those who received letermovir, there were reductions in csCMVi for the overall cohort and for high-risk patients at 100 days, 180 days, and 365 days after alloSCT. In seropositive recipients receiving seropositive alloSCT (CMV +/+), there were reductions in csCMVi at days 100, 180, and 365 after transplant. In seropositive recipients receiving seronegative alloSCT (CMV +/-), there was a reduction in csCMVi at 100 days after transplant. There was a reduction in CMV-related mortality at day 180 post-transplant (p=0.03) but not at day 365 (p=0.46). Additionally, for the overall cohort, the letermovir group showed worse OS from days 180-365 (HR 1.938, 95% CI 1.143-3.285). In CMV seronegative patients receiving seropositive SCT (CMV -/+), no difference in csCMVi was detected at days 100, 180, and 365. A landmark analysis of Day 100 IgG level effect on csCMVi incidence demonstrated that lower IgG level (<485) is associated with higher csCMVi rates after Day 100 (p=0.004). Day 100 CD4 level was not found to be predictive of csCMVi (p=0.744).
Conclusion: Letermovir prophylaxis reduces csCMVi overall and in high risk and CMV seropositive patients. We observed a reduction in CMV-related mortality at 180 days after transplant with letermovir prophylaxis. A worsening of OS after day 180 and a trend towards worse non-relapse mortality was surprising and may be related to a higher degree of immunosuppression associated with post-transplant cyclophosphamide (PTCy) use, haploidentical SCT, thymoglobulin use, prior SCTs, and CMV high-risk* patients. Additionally, letermovir was stopped at day 100 at which point survival worsens and the incidence of csCMVi precipitously increases. Our results suggest that there may be additional benefit to extending letermovir therapy past day 100 in the high-risk groups, haploidentical SCT group, patients with low IgG levels at day 100, and subclinical CMV viremia.
*High risk patients were defined as those receiving haploidentical, mismatched related, mismatched unrelated, umbilical cord blood, T-cell depleted graft stem cell transplants or immunosuppression with methylprednisolone > 1 mg/kg or 2 or more immunosuppressants.
Figure 1 Figure 1.
Disclosures
No relevant conflicts of interest to declare.
HLA-DP mismatch and CMV reactivation increase risk of aGVHD independently in recipients of allogeneic stem cell transplant
