Incidence and Risk Factors For Pneumocystis Jirovecii pneumonia (PCP) Following Haematopoetic Stem Cell Transplant (HSCT)

Serious PCP infections are known to develop in immuno-compromised individuals. The risk and risk factors in patients with retroviral infections is well established but the data in HSCT recipients is scanty. With use of newer methods of graft versus host disease (GVHD) prevention like alemtuzumab and reduced intensity transplants (RIC) this analysis was undertaken to establish the incidence and identify the risk factors for PCP infections in HSCT patients.

1129 patients (M: 732, F: 397; median age: 49yr., range: 14:73) undergoing 1149 HSCT procedures from 2002 to 2012 were analyzed (20 had more than 1 procerdure). HSCT type was autologous (n=786) or allogeneic (n=363) for malignacies [Ac. Leukaemia (n=235), chr. leukaemia (n=38), lymphoma (n=390), myeloma (n=424), aplastic anaemia (n=8) and other conditions (n= 34)]. Patients either had HSCT during this period or were alive post HSCT and at risk of PCP during the same time frame. Patients were evaluated for routine bacteriology, PCP and respiratory viruses using PCR if they had clinical signs and symptoms of respiratory infection. All patients received PCP prophylaxis (Co-trimoxazole prefereed; azithromycin, pentamidine and dapsone in cases intolerant to co-trimoxazole). Cummulative incidence rate was calculated using competing risk analysis with death as competing risk factor. Estimate of incidence was calculated per period of prophylactic drug used in cases where there was change in the medication (total 1930 spells in 1129 patients following 1149 procedures). Risk factors were analysed for the entire group and also separately for allograft and autograft population.

In allograft group, donor was sibling (n=162) or MUD (n=201). Conditioning was full intensity (n=154) or reduced intensity (n=210). Alemtuzumab was used in conditioning in 167 procedures, ATG in 23 and TBI in 208 cases (2Gy as RIC in 26). Stem cell source was PBSC (n=1050), BM (n=78), both (n=20) and 1 had cord blood. In 275/363 (76%) donor recipient pairs where CMV status was recorded, at least one of the pair was positive in 140/275 (51%) cases. 32 patients (2.8%) developed symptomatic, PCR positive PCP infection (5yr. risk 1.02%, 95% CI: 1.01-1.03%, SE: +0.004). In all but 4 cases PCP was the only identified pathogen (4 had respiratory virus in addition to PCP). Median time to develop PCP was 765 days (range: 70-2066 days). In univariate analysis, risk of PCP was not influenced by gender (M:24/1304 vs. F:8/626, p=0.33), age at HSCT (<49 yr.: 11/841 vs. >49yr.: 21/1089, p=0.21), CMV status of patient (Neg: 14/ 974 vs. Pos: 16/908, p=0.70), risk category of disease (Good: 24/1401, Poor: 8/438, p=0.33), or primary diagnosis (AL: 14/539, MM: 6/707, rest: 12/684, p=0.08). The risk was significantly higher with allograft (25/887 vs. 7/1043, p<0.0001), RIC (RIC: 22/541, Full: 3/350, Auto: 7/1039, p=0.0001) and use of alemtuzumab (19/388 vs. ATG: 1/69 vs. rest: 12/1473, p=0.0001). 11/537 episodes developed when patients were off PCP prophylaxis but risk was significantly higher during the period of azithromycin as PCP prophylaxis (17/321, vs. Co-trimoxazole: 3/844 vs. Pentamidine: 1/222, p=0.0001). Mutivariate analysis for the entire group showed Azithromycin use (HR: 1.6, 95% CI: 1.14-2.3, p=0.007) and RIC (HR: 3.1, 95% CI: 1.2-7.9, p=0.019) to be independently associated with risk of PCP. In autograft group, there was trend towards higher incidence with Azithromycin prophylaxis (HR: 1.74, 95% CI: 0.91-3.0, p=0.073) while in allograft group use of azithromycin (HR: 1.67, 95% CI: 1.1-2.5, p=0.013) and RIC (HR: 5.14, 95% CI: 1.2-22.1, p=0.028) were independently associated with higher risk of PCP. Maximum risk of reactivation was in patients receiving alemtuzumab conditioned RIC allograft during azithromycin prophylaxis. Out of the 11 who were off prophylaxis, 4 had stopped azithromycin in prior 6 weeks. There was no relation to lymphocyte recovery and PCP infection in allograft patients.

Even though the overall risk of PCP is small in HSCT patients, there is a significantly higher risk in alemtuzumab conditioned RIC allograft. Infections seem to develop late and azithromycin as prophylaxis may be sub-optimal in this group. Future studies need to determine the duration of PCP prophylaxis in RIC allografts.

Cavet:Lilly: Membership on an entity’s Board of Directors or advisory committees; Celgene: Consultancy, Honoraria.