Title : Low dose oral letermovir prophylaxis strategy for cytomegalovirus prevention after paediatric haploidentical hematopoietic stem cell transplantation with post-transplant cyclophosphamide
Abstract:
Introduction: Cytomegalovirus (CMV) reactivation remains a major complication after paediatric hematopoietic stem cell transplantation (HSCT) with Haploidentical (Haplo) related donor with post-transplant Cyclophosphamide (PTCy). Letermovir is effective for CMV prophylaxis, but it is costly. We describe here our experience of low dose letermovir prophylaxis strategy for CMV prevention in Haploidentical HSCT.
Objectives: We retrospectively analysed data of 54 consecutive children who underwent Haploidentical HSCT with PTCy receiving letermovir as prophylaxis for CMV (March 2025- Sep 2026) and compared with historical pre-letermovir cohort of 121 Haplo HSCT (2017-2025) recipients for CMV reactivation and other CMV related outcomes. The primary endpoint was treatment-requiring CMV breakthrough. Secondary endpoints included any CMV reactivation, peak viral load, CMV disease, rescue antiviral use, graft-versus-host disease (GVHD) and mortality.
Methods: During our study period only 240-mg and 480-mg oral film-coated tablet formulations were commercially available. We used half or Full tablet as appropriate to weight band, to minimize drug wastage while ensuring that patients received 50% of the recommended dose. Prophylaxis was started from day +1 post-HSCT and continued till day +90. CMV reactivation was defined as any detected CMV copy number by PCR in blood sample. Regular Monitoring for CMV PCR was done twice weekly till day+30 then once weekly post HSCT till day+90 Since letermovir is known to cause CMV blips (Transient, self-limiting brief low level of detected copy number). We had kept cut-off of CMV copy number >1500/ml to start CMV directed treatment. Administered letermovir was converted to daily-equivalent exposure and expressed as a percentage of the recommended dose after accounting for cyclosporine co-administration.
Results: Median age was 7.0 years (range 0.92-18), median weight 26.15 kg (range 9-98), and 35/54 (64.8%) were male. Major indications included thalassemia (46.3%), sickle cell disease (14.8%), bone marrow failure syndromes (13.0%), Acute leukaemia (10%), and inborn errors of immunity (3.7%). Cyclosporine was administered in 17/54 (31.5%). Letermovir dose as recommended daily dose equivalent (RDDE) was calculated: RDDE-12.5% given to 2 cases, RDDE-25% given to 32 Cases and RDDE-50% given to 19 cases. Thus 53/54 (98.1%) received ≤50% RDDE. Except one case who received full dose. Out of all cases, any CMV reactivation in letermovir cohot 24/54 (44.4%) versus 65/121 (53.7%) in historical cohort (RR 0.83, 95% CI 0.59-1.16; p=0.326). CMV reactivation in letermovir cohort (24/54) had Median peak CMV load was 1,855 copies/ml (range 270-790,000) out of which 11/24 (45.8%) remained <1,000 copies/ml and 13/24 (54.2%) reached ≥1,000 copies/ml. In letermovir cohort 15/24 (62.5%) patients had CMV blips, and required no treatment.CMV specific treatment was given in Letermovir cohort 9/54 (16.7%) versus 60/121 (49.5%) in historical cohort (RR 0.38, 95% CI 0.20-0.71; p=0.0055). Among patients with CMV reactivation, specific treatment was required in only 9/24 (37.5%) versus 60/65 (92.3%). CMV disease occurred in 2/54 (3.7%), both CMV colitis. In letermovir cohort Ganciclovir/valganciclovir and cidofovir were used in 6/54 (11.1%) and 5/54 (9.3%), respectively. Treatment-requiring breakthrough was similar with versus without cyclosporine (17.6% vs 16.2%; p=1.000), supporting interpretation against the cyclosporine-adjusted reference dose. Acute GvHD occurred in 11/54 (20.4%) Grade I–II occurred in 8/54 recipients (14.8%), while Grade III–IV aGvHD occurred in 3/54 (5.6%) and chronic GvHD occurred in 6/54 (11.1%), respectively. All-cause mortality was 9/54 (16.7%) versus 27/121 (22.3%) historically (RR 0.75, 95% CI 0.38-1.48; p=0.427).
Conclusion: Low dose letermovir prophylaxis (≤50% RDDE in 98.1%) significantly reduced treatment-requiring CMV breakthrough after paediatric haploidentical HSCT with PTCy compared with historical controls (16.7% vs. 49.5%; p=0.00055), Most reactivations (62.5%) required no rescue therapy. These findings support reduced-dose letermovir is effective strategy in resource-limited settings, warranting validation in larger prospective studies.

