Abstract
Introduction: Unrelated donor CB transplantation (CBT) is effective therapy for patients (pts) with high-risk hematologic malignancies without suitable adult donors. CBT, however, can be complicated by CMV infection in seropositive recipients with viremia rates as high as 100%. Moreover, CMV infection is associated with increased transplant-related mortality (TRM) in some reports. Therefore, our center is investigating: 1) the efficacy of pre-transplant ganciclovir (GAN) prophylaxis, 2) the efficacy of early pre-emptive therapy based on frequent serial qPCR monitoring, & 3) anti-viral therapy toxicities. The aim of this strategy is to rapidly eradicate viremia & thereby prevent disease without prohibitive toxicity.
Methods: Eligible pts were adult CMV seropositive 1st allograft CB recipients transplanted 3/2013-2/2016. Pts received Cy 50/Flu 150/Thio 5-10/TBI 400 cGy myeloablative conditioning + CSA/MMF (no ATG). GAN prophylaxis (5 mg/kg IV daily) was given days -7 to -2. Viremia was evaluated by qPCR (lower detection limit 137 copies/ml) twice weekly from day +14 or earlier if indicated. Treatment for viremia was usually initiated at 1st or 2nd qPCR positivity. Foscarnet (FOS), GAN or valganciclovir (VALGAN) dosing (maintenance vs induction) was selected based on viremia level & assessment of severe infection vs toxicity risk. CMV disease was treated with induction dosing. CMV response was defined as 3 consecutive negative qPCR & no evidence of disease.
Results: 42 CBT recipients [median 51 years (range 23-66), 33 acute leukemias, 6 MDS/MPD, 3 lymphomas] received double unit grafts [median donor-recipient HLA-allele match 5/8 (range 2-7/8) & infused CD34+ cell dose 1.3 (range 0.2-3.2) x 105/kg/unit]. 98% of pts engrafted (1 graft failure unrelated to CMV) & 86% (26 gr. II, 7 gr. III, 3 gr. IV) had gr. II-IV aGVHD by day 100 (median onset 29 days, range 19-35). Of the 42 pts, 35 reactivated CMV in the first 100 days [median viremia onset 33 days (range 5-74) & median viral load at 1st detection < 137 copies/mL (range < 137-245)] for a day 100 cumulative incidence of 83% (95%CI 67-92). Of 31 pts with both CMV & aGVHD, 15 developed viremia prior to & 16 after aGVHD onset. The median time from CMV detection to therapy initiation was 3 days (range 0-39). Treatment dosing, viremia vs disease & responses are shown (Figure). Induction dosing was given to one third of pts (n = 11). These pts started therapy later in their transplant course & predominantly received VALGAN. By day 100, just over half cleared viremia but none developed disease. Two-thirds of pts (n = 24) were started on maintenance dosing. They required therapy earlier post-CBT & most received FOS. 10 of these pts cleared viremia by day 100 although GI CMV disease developed in 1 pt after early therapy cessation. This pt was then successfully treated. Notably, half of the maintenance pts required escalation to induction due to persistent viremia in 10 or development of CMV pneumonia in 2 pts. The 2 pneumonia pts developed disease 26 & 12 days post-therapy initiation, respectively, & their peak viremias were 1,330 & 613 IU/ml, respectively. Both pts died of CMV & constituted 2/4 TRM deaths by day 100 in the study (other deaths were 1 DAH & 1 grade IV aGVHD). We then examined therapy toxicity in the first 100 days (Table). 4 pts received only FOS, 10 only GAN/ VALGAN, & 21 had courses of each. Therapy switches were due to toxicity, inadequate response, or convenience of oral VALGAN & therapy duration was highly variable according to infection severity, concurrent GVHD therapy, speed of viremia clearance & tolerance of drug toxicities. One-third of FOS pts & over half of GAN/ VALGAN pts had clinically significant toxicities.
Conclusion: CMV infection is very frequent in adult seropositive CBT recipients & pre-CBT GAN is not effective prophylaxis. Early intensive monitoring permits early pre-emptive therapy which is effective in most pts. While maintenance dosing is effective in some, the development of lethal CMV pneumonia in 2 maintenance pts suggests dose escalation may be appropriate if rapid viremia eradication is not achieved. However, FOS/GAN/VALGAN toxicities are also significant. Ultimately, the optimal drug dosing to enhance efficacy but mitigate toxicity, how to predict pts at greatest risk of disease, appropriate therapy duration, & how to predict recurrence risk are not known. Improved CMV prophylaxis & therapy for allograft recipients including CBT pts are needed.
Disclosures
No relevant conflicts of interest to declare.
Occult Talaromyces marneffei Infection Unveiled by the Novel Mp1p Antigen Detection Assay