No abstract

No abstract

Background: Invasive microbial infections remain a major cause of morbidity and mortality in hematopoietic stem cell transplant (HSCT) patients. Microbes residing in the GI tract are believed to be the source of mucosal barrier injury-associated blood stream infections (mBSI) in HSCT and cancer patients. Three host factors are critical for preventing microbial translocation from the gut: neutrophils, intact gut barrier function, and gut microbiota homeostasis. Currently there are no reliable biomarkers that can identify patients who will develop mBSIs. Objectives/Hypothesis: Since all HSCT patients develop severe neutropenia (ANC<100), we hypothesized that HSCT patients who develop mBSI will have more impaired gut barrier function and a greater imbalance in gut microbiota populations (as evidenced by depletion of beneficial obligate anaerobe commensals) compared to counterparts who do not develop mBSI. Methods: We designed a prospective, non-randomized, case control pilot study for HSCT patients < 20 years of age. Serum citrulline, which has been used as a surrogate to measure overall gut function, was obtained 1-2 times weekly. Bacterial genomic DNA was isolated from stool samples collected every 7 days. Amplicons (16S rRNA V4 variable region) were generated and sequenced using the Illumina MiSeq platform. Microbial taxonomic assignments were determined using QIIME software and further characterized into relative abundances of obligate (beneficial commensal) versus facultative (pathogenic) anaerobes. Clinical characteristics were recorded. Results: Of the 78 HSCT patients enrolled on the study, 27% (21 patients) developed BSI. Of the patients with confirmed blood stream infections, 16 patients were categorized as mBSI, and ultimately 11 patients were included in our analysis. 48 patients did not develop a BSI and had evaluable citrulline levels and constituted our control group. Patients with mBSI had a more significant decline in citrulline levels during the period from the start of conditioning to the development of mBSIs (linear mixed effect model, p = 0.048, likelihood ratio test) compared to counterparts without mBSI. We found no significant association between levels of facultative (pathogenic) anaerobes and the development of mBSI in this cohort. Conclusions: Decreased citrulline levels, which suggest impaired gut epithelial function, are associated with the development of severe, invasive bacterial infections. These data suggest that measuring citrulline may have utility as a biomarker for predicting mBSI in HSCT patients. Disclosures Koh: Merck: Consultancy.

418 Background: Patients with clinical stage I (CS I: cN0M0) germ cell tumors (GCT) exhibit favorable oncologic outcomes. While prognostic features can help inform treatment in adults with CS I GCT, we lack reliable means to predict relapse among pediatric patients. We sought to identify predictors of relapse in children with CS I GCT. Methods: We performed a pooled post hoc analysis on pediatric CS I GCT patients enrolled in 3 prospective trials: INT-0097 (phase II), INT-0106 (phase III), and AGCT0132 (phase III). Pathology was centrally reviewed. Patient demographics, pT stage, serum tumor markers, margin status, histology, relapse, and survival were compiled. Cox regression analyses were used to identify predictors of outcomes. Results: 88 patients were identified with histologic data available. Most patients were pT1-2 stage. Yolk sac tumor was present in 75%, while 16% had embryonal carcinoma, and 9% had choriocarcinoma. When evaluable, lymphovascular invasion (LVI) was present in 36/66 (55%) of patients. Over a median follow-up of 5.0 years, no patients died and 24 patients (27%) relapsed (median relapse-free survival not reached). Predictors of relapse included presence of choriocarcinoma (HR 4.3, p=0.004), embryonal carcinoma (HR 3.8, p=0.002), pT3 stage (HR 6.9, p=0.027), and age >12 years (HR 3.1, p=0.011). LVI (HR 2.4, p=0.072), serum tumor markers, and dominant tumor size did not reach significance. Pediatric CS I GCT patients exhibit remarkable 5-year survival. Conclusions: Using combined data from multiple prospective trials, our study identifies clinicopathologic features that predict relapse and potentially inform personalized treatment for these patients.

No abstract