Background: Low MBL2 concentration and MBL2 genotype variants have been associated with an increased risk of infection in various clinical settings. Pulmonary infection is a major complication of HSCT. We examined the relationship of MBL genotypes with post-engraftment bacterial (B-PNA) and fungal (F-PNA) pneumonia

Methods: Retrospective review of 236 non-consecutive, non-selected patients who underwent HSCT at MSKCC from 1/1/2000–4/30/2007. Microbiologically confirmed infections and pneumonias were recorded. Antifungal prophylaxis consisted of fluconazole 400 mg daily. Patients at high risk for mold infection received mold-active prophylaxis. After 1/1/2006 voriconazole was the first line anti-mold prophylaxis. Genotype was determined by PR-Melting Curve Analysis on blood or buccal swab specimens. MBL genotype was classified as wild-type: A/A (MBL-sufficient, MBL-S) or variant-type: A/O, O/O (MBL deficient, MBL-D). Patients were followed for up to 2 years. Statistical analysis: Fisher’s Exact test was used to compare the incidence rate between MBL-S and MBL-D patients. Multivariate logistical regression models were used to investigate the relationship between bacterial or fungal pneumonia and MBL genotype, matched related donor (MRD), myeloablative conditioning (MC) peripheral blood as stem cell source (PBSC), acute GVHD grade 2–4 (aGVHD). The results from Maximum Likelihood Estimates were summarized.

Results: Transplant characteristics: 80% MC, 76% PBSC, 48.6% MRD. Incidence of aGVHD: 22.4%. MBL genotypes: One-hundred and forty-two (60%) patients were homozygous for wild-type MBL2 (AA), 85(36%) were heterozygous (A/0) and 9 (3.8%) were homozygous for variant genotypes (OO). Transplant characteristics, rates of GVHD, relapse-free and overall survival were similar between MBL-D and MBL-S. There was higher incidence of overall bacterial infections in MBL-D compared to MBL-S pts (47.87% vs 36.62%, p=0.1049). MBL-D had a higher incidence of B-PNA (12.7% vs 4.9%, p=0.048). In multivariate logistic regression analysis, MBL-D(p=0.04) and aGVHD(p=0.06) were likely associated with B-PNA. Rates of overall fungal infections and F-PNA were similar [among MBL-D and MBL-S pts (12.77% vs 9.86%, p=0.5277) and (8.51% vs 7.75%, p=0.1049) respectively]. In multivariate logistic regression analysis only aGVHD was statistically significantly associated with F-PNA p=0.0002.

Conclusions: 1) MBL-D genotype was likely associated with increased risk of bacterial pneumonia. 2) MBL-D and aGVHD were risk factors for B-PNA. 3) Further analyses are in progress to evaluate the effect of MBL-D on fungal pneumonia in patients who received mold prophylaxis versus patients who did not receive mold prophylaxis. 4) Prospective studies are needed to assess the relative contribution of MBL-D genotype on the risk of pulmonary infection in HSCT.

Disclosures: Ding:Enzon: Employment, Equity Ownership. Kalambakas:Enzon: Employment, Equity Ownership. Buchbinder:Enzon: Employment, Equity Ownership.

Author notes

Corresponding author