Background :

Reactivation of cytomegalovirus (CMV) commonly occurs post allogeneic stem cell transplant (ASCT). The incidence is at least 40-50% in India, where most patients and donors are CMV seropositive. Ganciclovir, although the drug of choice, is expensive and associated with significant toxicities. Second line agents like foscarnet and cidofovir are not available in India leaving few therapeutic options. Leflunomide, an anti rheumatoid arthritis drug, acts by inhibiting virion assembly of CMV and has been effective in post renal transplant CMV infections. This drug is inexpensive and has a favourable toxicity profile. There have been only occasional case reports of leflunomide post ASCT. Here we report the efficacy of leflunomide in CMV reactivation in ASCT setting.

Methods :

All patients who underwent ASCT at our centre between April 2015 and February 2017 were included in this retrospective study. Serologic evaluation for CMV infections was done at baseline for all patients and donors. Conditioning regimens used were either myeloablative [total body irradiation (TBI)-cyclophosphamide (Cy)] or reduced intensity (fludarabine with either melphalan, or Cy or treosulfan +/- 2 gray TBI). Graft versus host disease (GVHD) prophylaxis used was cyclosporine or tacrolimus with methotrexate or mycofenolate mofetil for matched sibling donor (MSD) and matched unrelated donor (MUD) transplants. Post transplant Cy was used for haplo-identical transplants (HIT).

Monitoring was done twice weekly using quantitative CMV DNA polymerase chain reaction (QPCR) on peripheral blood post transplant till day +120 or longer if patient was on steroids. Patients who were detected to have a positive QPCR greater than 500 copies/mL at 2 consecutive time points were started on pre-emptive therapy. Ganciclovir was the preferred agent for pre-emptive therapy, however in case of ongoing cytopenias, leflunomide was used. In case of clinical resistance to ganciclovir, leflunomide was added as second line. Leflunomide was started at a loading dose of 100 mg once a day for 3 days followed by daily maintenance dose of 20 mg till clearance of CMV (undetectable CMV DNA on QPCR at 2 consecutive time points) or intolerance. Failure of leflunomide was defined as > 1 log increase in CMV copy number at any time or if another anti CMV agent was added due to CMV disease or non-clearance. Invasive procedures such as broncho-alveolar lavage, colonoscopic biopsies, and liver biopsies were used to detect CMV disease, when clinically suspected.

Results :

Seventy patients underwent ASCT during this period. This included 47 MSD transplants, 18 HIT and 5 MUD transplants. There were 49 episodes of CMV reactivation in 43 patients (61%). The incidence was highest in HIT with 17 of 18 patients developing reactivation.

Leflunomide was used in 24 (49%) of the 49 episodes of CMV reactivations. It was used as 1st line agent in 15 episodes and as 2nd line in 9. The median CMV copy number/ml was 0.82x 103 (range 0 to 13x103) on day of starting leflunomide. The median duration of use was 34 days (range 6 to 85 days). It was used pre-emptively in 20 episodes and as therapeutic (i.e. patient had end organ involvement) in 4. None of the patients had toxicity requiring stopping of the drug. Nine (38%) of the 24 CMV reactivation episodes cleared with leflunomide. When used as a 1st line agent, CMV infection cleared in 7 of the 15 episodes (46%) in comparison to ganciclovir as 1st line with which 12 (52%) of 23 episodes cleared.

Seven of the 24 episodes had initial CMV burden more than 2 x 103 copies/mL. Leflunomide failed in all of these, as against clearance in 9 of the 17 episodes which had CMV less than 2 x 103 copies/mL. (p= 0.022). None of the patients who received leflunomide in therapeutic setting cleared CMV (0 of 4), compared to 9 of 20 with pre-emptive leflunomide. (p= NS).


This is the largest study of using leflunomide for CMV reactivation in post ASCT setting to our knowledge. Leflunomide has considerable activity against CMV, however it is best used when CMV copies are less than 2000 /mL in pre-emptive setting. In therapeutic setting its role seems limited. A larger prospective study is warranted.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.