Febrile Neutropenia Admissions Research Articles

Outcomes of Children Discharged Prior to Absolute Neutrophil Count Recovery After Admission for Febrile Neutropenia.

Febrile neutropenia (FN) management in pediatric oncology patients traditionally necessitates inpatient admission until evidence of bone marrow recovery. Discharge before count recovery may be a way to safely reduce the length of hospitalizations for select patients. A chart review was conducted of patients admitted for FN at one tertiary care children's hospital, where the standard is to discharge well-appearing patients after 48 hours of negative cultures if afebrile for at least 24 hours, irrespective of absolute neutrophil count (ANC). Patients with ANC <500 at discharge were identified as early discharges, and data were collected with respect to rates of readmission and infectious complications in this cohort. Among 1230 FN encounters, 765 (62%) were early discharges. 122 patients (15.9%) were readmitted within 7 days. Patients with acute myeloid leukemia and ANC <100 at discharge were more likely to be readmitted. Of the early discharges, only 10 (1.31%) were readmitted with positive blood cultures and 5 (0.7%) were admitted to the pediatric intensive care unit within 24 hours of readmission. Routine discharge before ANC recovery allows for short hospital stays with low rates of readmission, infectious complications, and critical illness for pediatric oncology patients. This safe and beneficial policy should be considered at other institutions.

All-Cause Mortality and Its Predictors in Haemato-Oncology Patients with Febrile Neutropenia.

Febrile neutropenia (FN) is one of the most important life-threatening complications in haemato-oncology. Our objective was to report all-cause mortality rates in patients ill with a hematological malignancy (HM) hospitalized with a first FN episode and to identify predictors for mortality. We conducted a historical retrospective cohort study of consecutive patients with an HM, >18 years of age, admitted between January 2012 and August 2018 for a first episode of FN. Data on all-cause mortality 12 months after admission for FN were obtained. The Kaplan-Meier curve was used to describe mortality during the follow-up period. Univariate and multivariable analyses identified predictors for 1,3 and 12-month mortality. One hundred and fifty-eight patients (mean age 69.5, 49.4% males) were included. Overall, 54 patients died (15.8%, 25.9%, and 34.1% died after 1, 3, and 12 months, respectively). Lower serum albumin, higher serum gamma-glutamyl transferase (GGT), lower estimated glomerular filtration rate (eGFR), older age, higher temperature, and lower absolute lymphocyte count at admission were independent predictors of all-cause mortality after 12 months. Further studies are needed to confirm our results and identify therapeutic strategies to improve survival.

Identification of factors to predict low risk febrile neutropenia admissions in patients with acute myeloid leukemia (AML) following intensive chemotherapy.

7031 Background: Febrile neutropenia (FN) is the most common reason for hospital readmission following chemotherapy for AML. Standard practice for FN in AML is hospitalization for rapid administration of IV antibiotics, thus FN is a major driver of healthcare resource utilization and cost. Whether this approach is justified in all situations is unknown. While FN risk models have been developed focusing on solid tumors, factors predicting uncomplicated courses of FN in AML, which may allow identification of patients suitable for outpatient management, have not been studied. We therefore examined whether baseline clinical characteristics could predict which AML patients with FN have a lower risk of progression to severe illness. Methods: We retrospectively reviewed medical records from 7/29/2014-12/24/2022 to identify adults with FN (fever &gt;38.3C and white blood count &lt;1000/µL). Included were patients with high-grade myeloid neoplasm (&gt;10% blasts in blood/marrow) who received intensive chemotherapy (cytarabine dose &gt;=100 2 /day) within 8 weeks. We collected baseline clinical and disease variables. Outcomes were: infections identified, hospital length of stay (LOS), intensive care unit (ICU) admission, and survival. A “low-risk [LR]” outcome was defined as a LOS &lt;72hrs without ICU admission or inpatient death. Univariate and multivariable (MV) logistic regression models were used to assess covariate associations with outcomes. Results: We identified 350 FN admissions in 198 patients with a median age 62 (range: 21-87) yrs. 63% were newly diagnosed and 37% relapsed/refractory. 12% had prior transplant. Median duration of neutropenia prior to fever was 12 (1-287) days. 46% had transfusion &lt;24 hours prior to fever. The median LOS was 5 (1-56) days. 9% of patients went to ICU and 6% died inpatient.Overall, 15% of admissions were classified as LR. 57% had an infection identified. The table shows factors in MV analysis that predicted an infection identified or high-risk (HR) admission. Conclusions: Only 15% of AML admissions for FN were LR. Factors that predict HR outcomes in standard FN models such the MASCC score—e.g. age, prior fungal infection —were not predictive of HR outcomes in AML. As a next step, we will compare our model to other published models used to predict poor outcomes after FN, which have not been developed or validated in AML. [Table: see text]

Phase Ib/2 Study of Oral Decitabine/Cedazuridine (ASTX727) and Venetoclax in Combination with the Targeted Mutant IDH1 Inhibitor Ivosidenib or the Targeted Mutant IDH2 Inhibitor Enasidenib in IDH Mutated Acute Myeloid Leukemia

Background Isocitrate Dehydrogenase (IDH) 1 or 2 mutations occur in ~20% of acute myeloid leukemia (AML). Both venetoclax (VEN)-based and targeted IDH-inhibitor (IDHi) therapies are effective treatment options for IDH mutated AML in combination with hypomethylating agents (HMA). ASTX727 is an oral, fixed dose (35 mg/100 mg) combination of decitabine and cedazuridine approved for the treatment of myelodysplastic syndrome (MDS). Herein we report the interim results of the first all-oral triplet regimen of ASTX727 (day 1-5) + VEN (day 1-14) in combination with a targeted mutant IDH1i ivosidenib (IVO) or IDH2i enasidenib (ENA), for IDH mutated AML. Methods Eligible patients were > 18 years old with relapsed/refractory (R/R) IDH1 or IDH2 mutated AML, or newly diagnosed (ND) AML not eligible for intensive chemotherapy. For the R/R AML cohort, prior VEN, HMA or IDHi use was not exclusionary. The primary objectives were to determine safety and the recommended phase 2 dose (RP2D) of ASTX727 and VEN with ivosidenib (Arm A) or enasidenib (Arm B) [Phase 1b], and to determine the composite remission rate (CRc; CRh+CRi+CR) for both arms [Phase 2]. Results A total of 32 patients have enrolled, with 3 screen failures and 2 patients ongoing first cycle of therapy. There are currently 27 evaluable patients (arm A: 10, arm B: 17) with median follow up of 5.7 months. Median age at enrollment was 73 (50 - 81). 44% (n=12) had ND AML and 56% (n=15) had R/R AML. European LeukemiaNet (ELN) risk was intermediate or adverse in 52% and 44%, respectively. Patients received a median of 4 treatment cycles (ND: 5, R/R: 3). The CRc rate in the overall population was 72% (ND: 100%, R/R: 53%). Measurable residual disease (MRD) negative CRc by flow cytometry was achieved in 64% (ND: 83%, R/R: 42%). (Table 1) In the R/R cohort, 78% (n=11) patients had received prior treatment with either an HMA (azacitidine or decitabine), BCL2i and/or IDHi, with a median of 2 prior treatments. (Table 2) 5 patients did not have prior VEN exposure of which 4 (80%) achieved an MRD negative CRc. Four patients transitioned to stem cell transplant (ND: 2, R/R: 2). The most common non-hematologic grade 3/4 AEs in the overall population included hyperbilirubinemia (n=2, 7%), representing indirect bilirubin in enasidenib-treated patients and mucositis (n=2, 7%) with one considered a residual effect of prior cytoreduction. There were two patients with possible/probable differentiation syndrome which resolved with medical management (dexamethasone and diuresis). Two episodes of reversible grade 1 tumor lysis syndrome were also reported. 60-day mortality within the ND and R/R cohort was 8% (n=1) and 0%, respectively. The ND patient withdrew from the study and went home on hospice after an admission for febrile neutropenia. Conclusions The combination of ASTX727 + VEN with IDH1 or IDH2 inhibition appears to be an effective regimen for the treatment of IDH mutated AML with high response rates including MRD-negative CRc, most notably in de novo patients and R/R pts without prior VEN exposure. AEs are anticipated and tolerable. Enrollment to this study is ongoing. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal

Infectious complications and long-term outcomes in patients with diffuse large B-Cell lymphoma and diabetes mellitus

Febrile neutropenia (FN) is a major complication in patients with diffuse large B-Cell lymphoma (DLBCL). Diabetes mellitus (DM) has deleterious effects on the immune system resulting in an increased risk of infections. We evaluated patients with DLBCL who started frontline treatment with R-CHOP, and compared outcomes according to presence of DM comorbidity. Between 2013 and 2018, 218 patients with DLBCL were included. 46 patients (21%) had DM. Rate of admissions for FN was higher for patients with DM (0.7 vs. 0.46 admissions/patient, p = .016), also after age and gender-matched subgroup analysis (p = .004). Improved glycemic control during FN hospitalizations was associated with shorter hospitalizations. Metformin was associated with improved median overall survival in diabetic patients (89 vs. 64 months, p = .018). In conclusion, Patients with DLBCL and DM had higher rates of FN hospitalizations. Improved glycemic control during FN hospitalization was associated with shorter length of stay.

Real-world effectiveness of prophylactic granulocyte colony-stimulating factor (G-CSF) early (week 1) and late (weeks 2-3) in the cycle for the prevention of febrile neutropenia (FN) among patients (pts) with breast cancer (BC) after high FN–risk chemotherapy (chemo).

599 Background: G-CSF mitigates chemotherapy-induced neutropenia (CIN) and reduces FN risk. G-CSF moves the nadir of absolute neutrophil count (ANC) earlier (to week 1) in the chemo cycle and shortens nadir duration (Crawford, NEJM 1991), suggesting the potential for suboptimal CIN protection early (week 1) in the chemo cycle. The relative FN risk in week 1 vs. weeks 2-3 of the cycle with G-CSF is unknown and was analyzed compared with no G-CSF in the real-world setting with high FN risk chemo. Methods: Using a database of administrative claims representing 100% of fee-for-service Medicare, we analyzed BC pts who initiated docetaxel (T), doxorubicin (A), or cyclophosphamide (C) monotherapy or combination therapy between 01/01/2015 – 12/31/2019. Sample pts included adults aged ≥ 65 years with continuous coverage in Medicare Parts A, B, and D for 6 months before and 20 days after chemo initiation. Pts were categorized as receiving vs. not receiving G-CSF therapy within 3 days after chemo. Rate of FN events starting in week 1 vs. weeks 2-3 in cycle 1 was calculated. We defined FN as an inpatient admission with a primary or secondary diagnosis of neutropenia and measured the interval between chemo initiation and FN admission. Results: Among 18,788 Medicare beneficiaries with BC treated with T, A, and/or C, 72% received G-CSF therapy. More pts receiving G-CSF were treated with ≥ 2 of T, A, and/or C compared to pts who did not receive G-CSF (71% vs. 51%). Overall FN incidence in cycle 1 was significantly lower among pts receiving G-CSF (4.0%; n=546) compared to pts not receiving G-CSF (8.8%; n=462) (p&lt;0.0001). In pts with G-CSF, 81% (440/546) of all 1st-cycle FN events started in week 1 vs. 19% (106/546) in weeks 2-3. In pts not receiving G-CSF, the start of 1st-cycle FN events was more equally distributed: 41% (190/462) started in week 1 vs. 59% (272/462) in weeks 2-3. Results were robust to sensitivity analyses restricted to pts receiving ≥ 2 of T, A, and/or C. The rates of 1st-cycle FN events starting in weeks 1 and 2-3 with and without G-CSF following chemo initiation is shown below. Conclusions: Prophylactic G-CSF was highly effective for the prevention of FN in weeks 2-3, but relatively ineffective in week 1 of cycle 1 in the real-world setting, leaving pts largely unprotected during the first week. This represents an unmet medical need in week 1 of the cycle, despite use of G-CSF. Clinical trial information: NCT03294577. [Table: see text]

Outcomes of Patients with Indolent Lymphoma Treated with Bendamustine Plus Rituximab Compared to Rituximab Plus CVP or CHOP Chemoimmunotherapy in Ontario

BackgroundFor patients with symptomatic advanced stage indolent lymphoma, rituximab (R) was traditionally used alongside cyclophosphamide, vincristine, prednisone (CVP), or cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP). Bendamustine plus rituximab (BR) has been found to increase progression-free survival compared to R-CVP/CHOP, albeit with possible increased toxicity. Since 2013, bendamustine has been approved for public reimbursement for treatment of indolent lymphomas in Ontario, Canada, and has become the preferred regimen. We aimed to assess survival and toxicity of patients with indolent lymphoma treated with BR compared with R-CVP/CHOP in Ontario.MethodsWe conducted a retrospective cohort population-based study using administrative healthcare databases from Ontario, Canada. Patients were included if they had a diagnosis of indolent lymphoma and were treated with R-CVP/CHOP from 2005-2012, or with BR from 2013-2018. Patients with mantle cell lymphoma were excluded. Treatment groups were propensity score matched based on lymphoma subtype, age, sex, prior cancer diagnosis, time since diagnosis and rural or urban residence. The primary outcome of the study was overall survival (OS) from time of first treatment to death or end of study period. Secondary outcomes included toxicities during the first nine months of treatment, such as infections and secondary malignancies, associated resource utilization, and outcomes of patients treated with rituximab maintenance therapy. Cox regression models were used to assess factors associated with mortality.ResultsA propensity-matched cohort of 2032 patients treated with BR and 2032 treated with R-CVP/CHOP (1473 patients treated with R-CVP, 559 patients treated with R-CHOP) were included. The median age of patients was 65 years and 48% were female. A majority of patients had a diagnosis of follicular lymphoma (58.9% of patients treated with BR, 59.7% of patients treated with R-CVP/CHOP, p=0.59). Median follow-up time was 41 and 87 months for patients treated with BR and R-CVP/CHOP, respectively.BR was associated with significantly lower mortality compared to R-CVP/CHOP (HR 0.77, 95% CI 0.67-0.89, p<0.01). Five-year OS was 80% and 75% for patients treated with BR and R-CVP/CHOP, respectively (Figure 1). A total of 969 patients died during a follow-up period of seven years - 331 (16.3%) treated with BR and 638 (31.4%) treated with R-CVP/CHOP. Causes of death were available in 861 patients (88.9%), shown in Table 1. Most patients in both groups died from lymphoma (66.9% treated with BR, 66.4% treated with R-CVP/CHOP, p=0.88). A Charlson comorbidity index greater than or equal to 2 was associated with increased mortality (HR 3.04, 95% CI 2.45-3.77).Treatment toxicities are listed in Table 2. Over a nine-month period after treatment initiation (induction), patients treated with BR had a mean of 0.85 emergency department visits compared to a mean of 1.01 for patients treated with R-CVP/CHOP (p<0.01). Admissions for febrile neutropenia were more common in patients treated with R-CVP/CHOP compared to BR (4.9% vs. 2.2%, p<0.01), whereas infections resulting in hospitalization were more common in patients treated with BR. (21.9% vs. 17.3%, p<0.01).A majority of patients in both cohorts received maintenance therapy (81% of patients treated with BR, 64.1% of patients treated with R-CVP/CHOP). The use of maintenance therapy with either regimen was associated with less mortality (HR 0.16, 95% CI 0.14-0.19). Among patients who received maintenance therapy, patients previously treated with BR had more hospital admissions for fever (21.5% vs. 18.3%, p=0.03) and infections (37.9% vs. 31.5%, p<0.01) over a three-year period.ConclusionIn these well-matched cohorts of patients with indolent lymphoma, improved OS was seen with BR compared to R-CVP/CHOP. Patients treated with BR had more infectious complications, although more episodes of febrile neutropenia were seen in patients treated with R-CVP/CHOP. The use of maintenance therapy was associated with less mortality, although patients who received maintenance therapy may have been a more fit and selected population. Higher infectious toxicities were seen for patients who received maintenance rituximab after BR. This study supports BR as the standard of care, and future prospective studies should evaluate the survival and toxicity impact of rituximab maintenance. [Display omitted] DisclosuresPrica: Astra-Zeneca: Honoraria; Kite Gilead: Honoraria.

ABCL-233: Febrile Neutropenia in Patients with Aggressive Large Cell Lymphoma and Diabetes Mellitus

<h3>Context</h3> One of the major complications in patients with aggressive lymphoma receiving intensive chemo-immunotherapy is febrile neutropenia (FN). Diabetes mellitus (DM) has deleterious effects on the immune system, resulting in an increased risk for infections, as well as a complicated course of infections. <h3>Objectives</h3> We aimed to examine whether comorbidity with DM, as well as diabetic control, affects the risk of FN or complicates the course of FN in aggressive lymphoma patients. <h3>Design, Setting, and Patients</h3> This is a retrospective, single-center study. We included all consecutive patients diagnosed with aggressive large cell lymphoma treated at our institute between 2013 and 2017. Demographics, disease-related outcomes, and laboratory data were extracted from electronic medical records. Categorical variables were compared using the ײ or the Fisher's exact test, and medians were compared with the Mann–Whitney U test. <h3>Results</h3> 226 patients with aggressive large cell lymphoma were treated in our medical center. Of these, 49 patients (22%) had type 2 DM. There were 129 hospital admissions due to FN in the entire cohort during lymphoma treatment. Patients with DM had higher mean admission rates due to FN compared to patients without DM (0.88 and 0.51 FN admissions per patient, respectively, P=0.004). The admission rate in patients with DM remained higher after age and gender-matched subgroup analysis (P = 0.005). However, the duration of hospitalizations and number of admissions to the ICU were similar in patients with and without DM. In patients with DM, improved glycemic control during hospitalizations due to FN was associated with shorter hospital stay. However, baseline hemoglobin A1c levels were not associated with increased risk for FN or infectious complications. Use of metformin was associated with an improved overall survival in diabetic patients (overall survival not reached compared to 43 months). <h3>Conclusions</h3> Patients with aggressive large cell lymphoma and DM had higher rates of hospitalizations due to FN. Furthermore, our data suggest that improved glycemic control during hospitalization is associated with a shorter length of hospitalization for FN. The positive effects of metformin in our study cohort may warrant further investigation.

Bacteremia in Pediatric Patients with Benign Neutropenia Presenting with Fever

Introduction: Benign neutropenia (BN) is a heterogeneous group of rare disorders, as compared to neutropenia in patients with underlying malignancy. Febrile neutropenia in children with BN is a frequent cause for admission for intravenous antibiotic administration due to the possibility of life-threatening infections. However, the majority of children diagnosed with BN experience a benign clinical course and may not require aggressive medical management each time they develop a fever. Scant data exist regarding the incidence of bacteremia in BN patients. The aim of this study was to assess the rates of bacteremia and other morbidities in a large patient population of children with BN.Methods: With IRB approval, the Pediatric Health Information System (PHIS) database was queried for children < 18 years admitted with fever and neutropenia between January 2011 and October 2015. The PHIS database captures de-identified patient information from 45 pediatric hospitals in the US. Patients were considered to have BN when identified with ICD9 codes for chronic, congenital, cyclic, genetic and immune neutropenia, and if they did not have a malignancy. Patients with neutropenia and underlying oncological diagnoses and who underwent HSCT served as controls. Bacteremia was identified using the ICD-9 diagnosis code (790.7). Infectious complications such as pneumonia, UTI and sepsis were also identified using ICD-9 diagnostic codes. The primary outcomes were admission rates, length of stay (LOS) and days of antibiotic use. The presence of bacteremia and other infectious complications was described with frequencies and per cents for each group and compared using Chi squared analysis or Fisher's exact test as appropriate. Continuous variables are described with medians and IQR and compared using Wilcoxon rank sum tests. LOS and days of antibiotic use were significant at the < 0.001 level. (Results are summarized in Table 1).Results: There were a total of 29,720 admissions for febrile neutropenia: 24,603 of patients with malignancies, and 5,117 of patients with benign neutropenia. There were 16,445 unique patients, 12,708 with malignancies, and 3,737 without. Fifty three percent (6, 783) of the patients with malignancies had only one admission captured at a PHIS center during the study period, while 82% (3,076) of the benign neutropenia patients had only 1 admission. From the 24,603 admissions of patients who had malignancy and fever and neutropenia, 1,488 (6.05 %) occurrences of bacteremia were identified. There were 168 (3.28%) occurrences of bacteremia identified in 5,117 admissions of patients with benign neutropenia and fever. Pneumonia occurrences were identified as follows: in 818 (3.32%) admissions of patients with malignancy vs 123 (2.4%) admissions of patients with benign neutropenia. Sepsis was identified as follows: in 685 (2.78%) admissions of patients with malignancy vs 95 (1.86%) admissions of patients with benign neutropenia. UTI occurrences were identified as follows: in 485 (1.97%) admissions of patients with malignancy vs 131 (2.56%) admissions of patients with benign neutropenia. Median LOS for patients with benign fever and neutropenia was 4 (3,7) days compared to 6 (4,12) days for patients with malignancies. Median days of antibiotic use was 3 (3, 6) days in benign neutropenia patients vs 5 (3, 10) days in patients with malignancy.Conclusions: In this large, multicenter database study we found very low rates of bacteremia both in patients admitted with fever and benign neutropenia and with neutropenia and underlying malignancy. The identified rates of pneumonia, sepsis and UTI were very low in both groups. The identified rates of pneumonia and sepsis were lower in patients with benign neutropenia and rates of UTIs were slightly higher in this group. Shorter LOS and shorter antibiotic courses were noted in patients with benign neutropenia compared with cancer patients. Further studies should explore the feasibility of outpatient management for febrile neutropenia in children with BN given the very low rates of bacteremia and infectious complications. [Display omitted] DisclosuresAhuja:Bayer: Honoraria; Shire: Honoraria; Shire: Speakers Bureau; CSL Behring: Honoraria.

A Retrospective audit comparing the neutropenic complications in the Pre-COVID and COVID era from a tertiary care center.

e18619 Background: The COVID era has shown a rise in Intensive Care Unit (ICU) admissions. This is mainly due to COVID related complications. On the contrary, the public around the world have become more aware about maintaining proper hygiene by wearing masks , using hand wash and keeping themselves clean. As an audit, we conducted a retrospective analysis comparing neutropenic complications during the pre COVID and COVID times at our tertiary care cancer hospital. Methods: We collected data of febrile neutropenia admissions in ICU from March to July 2019-Non Covid (NC) and March to July 2020-Covid (C). The data collection was done including data on nadir total WBC count, use of GCSF, positive cultures, type of malignancy and deaths. The statistical analysis was done using SPSS software version 22. Results: There were 15 patients in the NC dataset and 18 patients in the C dataset. The mean age of NC was 59 (SD-9.2) and that of C was 57. 3 (SD-14). The age and sex were comparable in both the groups. The number of solid organ malignancies constituted 5 (33.33%) in NC set and 8 (44.44%) in C set. The hematological malignancies constituted 10 in each of C and NC sets. Gram negative sepsis was seen in 26.67% and 22.22% in NC and C dataset respectively. There were 4 (26.67%) deaths in NC group and 5 (27.28%) C group. Growth factors were received by 9 (60%) patients in NC group and 15 (83.3%) patients in the C group (p=0.0005). There was no significant difference in the admission, complication and death rates in ICU when comparing the two groups. These results have been depicted in Table. Conclusions: Despite the precautions taken by community during the COVID times, we conclude that there was no significant difference in the neutropenic complications. It also suggests that most of the neutropenic infections are related to the person’s immunity rather than community transmission. The study limitation is that this is a single center study but further audits and prospective studies are warranted in this regard. Keywords - COVID, Non COVID, Neutropenia, Intensive Care Unit. [Table: see text]

Assessing Safe Discharge Criteria for Pediatric Oncology Patients Admitted for Febrile Neutropenia.

Recent studies suggest outpatient therapy, oral antibiotics, or earlier discharge could be appropriate in some pediatric patients admitted with febrile neutropenia; supporting data are lacking. Retrospective chart review of patients admitted from September 2005 through October 2016 identified 131 "early discharge" febrile neutropenia admissions with discharge absolute neutrophil count (ANC) <500/µl and negative cultures. All were afebrile and discharged without outpatient antibiotics. Eleven of 131 patients (8%) were readmitted. Two patients called back for late positive cultures. Nine were readmitted with febrile neutropenia; 2 had positive cultures on readmission. All 4 patients with positive cultures were safely treated with appropriate antibiotics. The remaining 7 patients had uneventful readmissions. Average ANC (SD) at discharge was lower for patients readmitted versus those not readmitted (69 [70] vs. 196 [145], P≤0.001), as was absolute phagocyte count (APC) at discharge (97 [82] vs. 453 [431], P≤0.001). APC on admission was not significantly lower for those readmitted (165 [254] vs. 321 [388], P=0.09). Few patients required readmission; those with bacterial infections were easily identified and appropriately treated. Higher ANC or APC criteria for discharge would increase length of hospital stay without decreasing morbidity. A subset of patients admitted with febrile neutropenia can be safely discharged before count recovery without oral antibiotics.

Incorporating Absolute Phagocyte Count With Absolute Neutrophil Count as a Measure for Safe Discharge for Pediatric Oncology Febrile Neutropenia: A Pilot Study.

Adequate bone marrow recovery is a discharge requirement after admission for febrile neutropenia in oncology patients, without specific threshold in consensus guidelines. In January 2016, our institution implemented count recovery criteria of absolute neutrophil count ≥100 cells/μL and absolute phagocyte count ≥300 cells/μL compared with prior criteria of absolute neutrophil count ≥500 cells/μL. Retrospective analysis comparing pre (July 2013 to December 2015, N=68) and post (January 2016 to June 2018, N=30) groups showed no difference in readmissions (P>0.9), no patient deaths, and decreased average length of stay in the post group (P<0.0001). Updated count recovery criteria seem feasible and safe.

Protecting high-risk oncology patients during the pandemic of COVID-19 by creating an outpatient cancer clinic for febrile neutropenia (California clinic).

191 Background: Delivering care for vulnerable cancer patients during a pandemic is challenging given the competing risks of death from cancer versus the high case fatality rates from SARS-COV-2 (CV-19). Data currently available suggests a total fatality rate close to 30%-50% with CV-19 in active malignancy patients. In addition to adapting guidelines from national organizations to reduce the social footprint of patients in order to minimize risk of exposure of CV-19, our cancer center implemented an isolated clinic with personal protective equipment (PPE) and direct access to a CV-19 rule out floor (if admission warranted) in order to manage those with febrile neutropenia (FN) who otherwise would have been triaged to the emergency room (ED). Methods: We implemented an outpatient isolated extended hour clinic with access to PPE, blood work, intravenous antibiotics and fluids for FN patients as a pilot project from mid-April with expected duration during the pandemic with the aim to decrease the ED admissions for FN by 50%. We used the Multinational Association of Support Care in Cancer (MASCC) validated tool to assist with outpatient versus inpatient management of these patients. All patients were screened via polymerase chain reaction nasal swab for CV-19 to identify CV-19 in a high-risk population. Our PDSA (Plan Do Study Act) cycles have been in 2-week sessions with constant re-education to multiple providers. Results: Prior to CV-19, our databases show an approximate 15 to 20 FN hematology and oncology patients per month who are triaged to ED during the business hours. Since the implementation of our clinic in the last 45 days, we have screened 8 patients, of which 2 were discharged home with oral antibiotics on isolation until CV-19 testing returned, 6 were directly admitted to CV-19 rule out floor avoiding ED. Our overall patient numbers were low during the peak of the pandemic and we expect to see increasing number of patients utilizing the clinic over the next few months. Conclusions: Implementing the California clinic has thus far successfully decreased the social footprint of our highest-risk cancer patients, those with FN, in hopes of decreasing their possible exposure to CV-19 as well as the unnecessary exposure of the clinical personnel. [Table: see text]

Impact of respiratory viral panel testing on length of stay in pediatric cancer patients admitted with fever and neutropenia.

Polymerase chain reaction (PCR) respiratory viral panel (RVP) testing is often used in evaluation of pediatric cancer patients with febrile neutropenia (FN), but correlation with adverse outcomes has not been well characterized. A retrospective cohort of all children ages 0-21years with cancer admitted to Children's Healthcare of Atlanta for FN from January 2013 to June 2016 was identified. Patient demographic and clinical variables such as age, RVP results, length of stay (LOS), and deaths were abstracted. Relationship between RVP testing and positivity and LOS, highest temperature (Tmax), hypotension and intensive care unit (ICU) admission were compared using Wilcoxon rank sums, chi-square, or Fisher's exact tests adjusting for age, sex, bacteremia, and diagnosis. The 404 patients identified had 787 total FN admissions. RVPs were sent in 38% of admissions and were positive in 59%. Patients with RVPs sent were younger (median 5.5 vs 8.0years, P<.0001) with higher Tmax (39.2° vs 39.1°, P=.016). The most common virus identified was rhinovirus/Enterovirus (61%). There were no significant differences in highest temperature or lowest blood pressure based on RVP positivity. Patients admitted to the ICU were more likely to have RVPs sent (odds ratio [OR]=3.19, P<.002); however, neither having RVP testing nor RVP positivity were significantly associated with increased LOS or death. Coinfection with bacteremia and a respiratory virus was identified in 9.1% of patients. These data raise the question of the utility of sending potentially costly RVP testing as RVP positivity during febrile neutropenia does not impact LOS, degree of hypotension, or ICU admission.

AML-225: Identification of Predictors for Uncomplicated, Low-Risk Febrile Neutropenia (FN) Admissions for Patients with Acute Myeloid Leukemia (AML) Following Intensive Chemotherapy

Context Prolonged hospitalizations for AML patients impair quality of life and drive leukemia care costs. Febrile neutropenia (FN) is the most common reason for re-hospitalization in AML as current algorithms call for rapid administration of IV antimicrobials and close monitoring. Whether this approach is justified in all situations is unclear given availability of novel oral broad-spectrum antimicrobials and home drug infusion services. While FN risk models have been developed focusing on solid tumors, factors predicting uncomplicated courses of FN in AML patients, which might allow the identification of patients suitable for outpatient management, have not been studied. Objective To identify factors to that define a low-risk population of AML patients admitted for FN. Methods We retrospectively reviewed medical records for adults admitted for FN (temperature ≥38.3°C and absolute neutrophil count 1 admission) assessed covariate associations with discharge. Results We identified 250 FN admissions in 129 AML patients from 7/29/14–6/26/19. Thirty-three (13%) admissions were Conclusions Only 13% of AML admissions for FN were short and uncomplicated. Factors that predicted more severe outcomes in solid malignancies did not individually predict such outcomes for AML patients; however, a model of combined risk factors resulted in a moderate-high predictive ability for a “low-risk” admission. Funding This work was made possible by generous support from the James Chung Yam Lee Pilot Award.

Time to Antibiotic for Pediatric Oncology Patients With Febrile Neutropenia at Regional Emergency Departments.

We compared the time to antibiotic (TTA) for pediatric oncology patients with febrile neutropenia (FN) presenting at regional emergency departments (EDs) with those presenting at a pediatric referral ED, and examined its association with need for aggressive medical care. We abstracted data for pediatric oncology patients (age, <21 years) admitted for FN between August 2012 and August 2017 at a single children's hospital and compared the TTA between those referred from a regional ED across the state and those admitted via the referral ED at the children's hospital. Factors associated with delay in antibiotic administration (TTA, >60 minutes) were estimated using generalized linear modeling with generalized estimating equations (GEEs). Delay in antibiotic administration was examined for its association with the need for aggressive medical care (>1 fluid bolus, intensive care unit admission, inotropic or invasive ventilator support) within 24 hours of admission as an exploratory aim. Three-hundred eighty-nine FN admissions (regional ED, 26.7%; referral ED, 73.3%) occurred in 205 eligible patients. Median TTA was significantly (P < 0.0001) greater among patients presenting at a regional ED (117.5 minutes [range, 9-722 minutes]) vs referral ED (46 minutes [range, 6-378 minutes]). Presentation at regional ED was the only factor associated with delay in antibiotic administration (odds ratio, 9.73; 95% confidence interval, 5.37-17.63; P < 0.0001). Delay in antibiotic administration was not associated with greater need for aggressive medical care (odds ratio, 1.34; 95% confidence interval, 0.55-3.29; P = 0.5). Pediatric oncology patients with FN presenting to regional EDs have longer TTA as compared with those presenting to a referral ED at a children's hospital.

Prophylactic Use of Filgrastim on Days 7, 11 and 14 in Patients Receiving R-CHOP for Non-Hodgkin's Lymphoma

Introduction Chemotherapy-induced febrile neutropenia (FN) is a side effect of cytotoxic chemotherapy and a major risk factor for infection as well as a dose-limiting toxicity. One method of reducing the incidence of febrile neutropenia is through the use of granulocyte stimulating factor (G-CSF). Patients with Non-Hodgkin's Lymphoma (NHL) receiving R- CHOP chemotherapy are thought to have a 10-20% risk of FN (Aapro et al 2010). Elderly patients (aged 65 and over) have an elevated risk of FN. The Newcastle upon Tyne Hospitals (NuTH) local guidelines recommend a novel regime prescribing one dose of filgrastim on days 7, 11 and 14 of treatment regime as primary prophylaxis. Several randomised studies and retrospective reviews of practice have examined the question of G-CSF prophylaxis in older patients. There is little doubt that the use of G-CSF as primary prophylaxis in this group reduces the incidence of FN, however the data are conflicting with respect to impact on overall survival (Osby et al 2003. Doorduijn et al 2003) Aims To audit prophylactic G-CSF prescribing against trust guidelines for patients over 65 with NHL receiving R-CHOP chemotherapy. We also assessed the frequency of hospital admissions for FN in patients over 65 receiving curative R-CHOP for NHL. Methods Patients with NHL over 65 who received R-CHOP at NuTH between 01/06/2017 - 31/01/2018 were identified using our electronic chemotherapy prescribing system. Electronic medical records were used to extract patient, chemotherapy, and hospital admission details. A data collection tool was used to record; patient age, cycle number, details about neutropenic sepsis related admissions and neutrophil counts between 1- 4 days prior to a cycle commencing. Results 30 patients with an average age of 73, cumulatively received 138 cycles of R-CHOP during the defined time period. 91.3% (126/138) of G-CSF prescriptions co-prescribed with chemotherapy were in accordance with trust policy. G-CSF on day 7, 11 and 14 corresponded with 8 hospital neutropenic admissions and 0 non-neutropenic. Comparatively 8.7% (12/138) of G-CSF prescriptions were for ≥5 days and corresponded with 5 hospital admissions with FN and 1 non-neutropenic. However, patients who received 5 or more days G-CSF had multiple risk factors for FN or had a previous episode of FN. The average length of admission due to FN was 6 days which has an associated financial implication. 17 chemotherapy prescriptions were dose reduced, of which 82.4% (14/17) were co-prescribed with the day 7, 11 and 14 regime. There were no deaths as a result of neutropenic sepsis in this cohort of 30 patients. Conclusion The NuTH G-CSF regime resulted in FN admission at a rate of 6.35% (8/126). This is lower than the expected rate of 10-20 % in the absence of G-CSF prophylaxis. These data support the use of the modified regimen of filgrastim on days 7, 11 and 14. References Aapro, M.S. et al. 2010 update of EORTC guidelines for the use of granulocyte-colony stimulating factor to reduce the incidence of chemotherapy-induced febrile neutropenia in adult patients with lymphoproliferative disorders and solid tumours. European Journal of Cancer, Volume 47, Issue 1, 8 - 32 Doorduijn JK, an der Holt B, Imhoff GW, Hem KG, Kramer MHH, Oers MHJ, et al. CHOP compared to CHOP plus granulocyte colony‐stimulating factor in elderly patients with aggressive Non‐Hodgkin's Lymphoma. Journal of Clinical Oncology 2003;21(16):3041‐50. Ösby K, Hagberg H, Kvaloy S, Teerenhovi L, Anderson H, Cavallin‐Stahl E, et al. CHOP is superior to CNOP in elderly patients with aggressive lymphoma while outcome is unaffected by filgrastim treatment: results of a Nordic Lymphoma Group randomized trial. Blood 2003;101(10):3840‐8. Disclosures Gabriel: Abbvie: Consultancy, Speakers Bureau; Jazz: Consultancy, Speakers Bureau; Kite: Honoraria. OffLabel Disclosure: Filgrastim - granulocyte colony stimulating factor

Limiting Vancomycin Exposure in Pediatric Oncology Patients With Febrile Neutropenia May Be Associated With Decreased Vancomycin-Resistant Enterococcus Incidence.

Limited data exists regarding the effects of empiric antibiotic use in pediatric oncology patients with febrile neutropenia (FN) on the development of antibiotic resistance. We evaluated the impact of a change in our empiric FN guideline limiting vancomycin exposure on the development of vancomycin-resistant Enterococcus in pediatric oncology patients. Retrospective, quasi-experimental, single-center study using interrupted timeseries analysis in oncology patients aged ≤18 years with at least 1 admission for FN between 2009 and 2015. Risk strata incorporated diagnosis, chemotherapy phase, Down syndrome, septic shock, and typhlitis. Microbiologic data and inpatient antibiotic use were obtained by chart review. Segmented Poisson regression was used to compare VRE incidence and antibiotic days of therapy (DOT) before and after the intervention. We identified 285 patients with 697 FN episodes pre-intervention and 309 patients with 691 FN episodes postintervention. The proportion of high-risk episodes was similar in both periods (49% vs 48%). Empiric vancomycin DOT/1000 FN days decreased from 315 pre-intervention to 164 post-intervention (P < .01) in high-risk episodes and from 199 to 115 in standard risk episodes (P < .01). Incidence of VRE/1000 patient-days decreased significantly from 2.53 pre-intervention to 0.90 post-intervention (incidence rate ratio, 0.14; 95% confidence interval, 0.04-0.47; P = .002). A FN guideline limiting empiric vancomycin exposure was associated with a decreased incidence of VRE among pediatric oncology patients. Antimicrobial stewardship interventions are feasible in immunocompromised patients and can impact antibiotic resistance.

Re-audit of hospital admissions for febrile neutropenia (FN) in patients receiving carboplatin and etoposide chemotherapy

Re-audit of hospital admissions for febrile neutropenia (FN) in patients receiving carboplatin and etoposide chemotherapy

Clinical and Laboratory Benefits of Early Initiation of Hydroxyurea with Pharmacokinetic Guided Dosing for Young Children with Sickle Cell Anemia

Background: Hydroxyurea is now the standard of care for children with sickle cell anemia (SCA). Results from the BABY HUG study and recommendations from the 2014 NHLBI Guidelines have led to early initiation (increasingly before 1 year of age) of hydroxyurea for many patients. Given the known variability in hydroxyurea pharmacokinetics (PK), treatment response (HbF%), and maximum tolerated dose (MTD), we hypothesized that individualized dosing would provide the optimal treatment approach.Methods: The Therapeutic Response Evaluation and Adherence Trial (TREAT, ClinicalTrials.gov NCT02286154) is a prospective study of a personalized, PK-guided dosing model of hydroxyurea for children with SCA. Using population PK model-based Bayesian estimation, each participant's PK data are used to generate an individualized starting hydroxyurea dose that targets an area under the curve associated with actual MTD. Clinical follow-up and subsequent dose adjustments target MTD, usually defined by ANC<3.0x109/L. We analyzed clinical and laboratory data for TREAT participants who started hydroxyurea before 2 years of age, to allow for comparison to published results from BABY HUG, which included a similar young cohort but with conservative weight-based dosing of 20 mg/kg/day. TREAT participants had ongoing clinical and research evaluations of organ function, including transcranial doppler (TCD) studies, RBC pit counts, and cystatin C measurements.Results:The analysis of children starting hydroxyurea before 2 years of age included 33 participants (of 47 total TREAT enrollments), who contributed a total of 59.5 patient-years of hydroxyurea therapy. The mean age (±SD) at hydroxyurea initiation was 1.0±0.4 years of age. The average PK-guided, individualized starting dose was 27.8±5.3 mg/kg/day, higher than conventional and BABY HUG initial dosing (20 mg/kg/day). For children who have completed 12 months of therapy (n=24), effects in hematologic laboratory data are remarkable with average 35.9±8.9% HbF and hemoglobin concentration of 10.2±1.1 g/dL after 12 months of therapy (compared to 29.3±8.8% and 9.2±1.3 g/dL at baseline). The majority (70%) of these participants have HbF>30% and almost half achieved HbF>40% after 12 months of hydroxyurea. This hematological response is more robust than what was observed in BABY HUG (HbF=22.4%, Hb=9.1 g/dL after two years of therapy, Wang WC et al. Lancet 2011). In the TREAT cohort, there were no episodes of dactylitis, acute splenic sequestration, or stroke. There were 111 emergency room or sick outpatient clinic visits for this young cohort; 107 ED/clinic visits (without subsequent hospitalization) were for fever, URI symptoms, GI illness, or other non-specific complaints unrelated to SCA, while only 4 (3.6%) visits were for pain. There were 38 hospitalizations in 17 participants, mostly for routine evaluation of fever (66%), but no positive blood cultures and no admissions for febrile neutropenia. The average length of hospitalization was 2.8±2.4 days with 81% of participants discharged within 72 hours of admission. There were 3 episodes of acute chest syndrome in 2 patients, two of whom required PRBC transfusion. Including all types of visits, there were only 6 pain events, equivalent to 10.1 pain events per 100 patient-years, which is much lower than the published 94 events per 100 patient-years in the hydroxyurea treatment arm of BABY HUG (Thornburg CD et al. Blood 2012). There were 37 TCD exams performed in 16 participants, all normal except for one patient with conditional velocities that normalized with hydroxyurea. There were no significant differences from baseline to month 12 in either RBC pit counts or cystatin C values.Conclusions: Hydroxyurea initiation at an early age using PK-guided dosing provides significant clinical benefits for young children with sickle cell anemia. These TREAT study data suggest that initiating hydroxyurea around one year of life using a personalized dosing strategy can provide better clinical and laboratory benefits than starting at the conventional 20 mg/kg/day weight-based dose. Very high HbF levels are observed at modest and well-tolerated doses of hydroxyurea, perhaps because treatment was initiated before the process of HbF inactivation is complete. Continued long-term follow-up of these patients will determine whether these will be sustained and able to prevent both short- and long-term complications of SCA. DisclosuresMalik:CSL Behring: Patents & Royalties. Quinn:Silver Lake Research Corporation: Research Funding; Global Blood Therapeutics: Research Funding; Amgen: Research Funding. Ware:Biomedomics: Research Funding; Nova Laboratories: Consultancy; Bristol Myers Squibb: Research Funding; Addmedica: Research Funding; Global Blood Therapeutics: Other: advisory board; Agios: Other: advisory board; Novartis: Membership on an entity's Board of Directors or advisory committees.