About

Athena F. Zuppa, MD, MSCE, is an Adjunct Professor of Pediatrics (General Pediatrics) at the University of Pennsylvania's Perelman School of Medicine. She completed her undergraduate studies in Chemistry at Colgate University in 1992, earned her MD from the State University of New York at Stony Brook in 1996, and obtained a Master of Science in Clinical Epidemiology and Biostatistics (MSCE) from the University of Pennsylvania in 2003. Dr. Zuppa's professional focus includes clinical epidemiology and biostatistics within the field of pediatrics, contributing to patient care and medical research at Penn Medicine.

Research topics

• Medicine
• Anesthesia
• Internal medicine
• Intensive care medicine
• Chemistry

Selected publications

• Camelot-2: A phase 3 Randomized, Double-Blind, Placebo-Controlled, Study of Bleximenib, Venetoclax and Azacitidine for the Treatment of Participants with Newly Diagnosed Acute Myeloid Leukemia Harboring KMT2A rearrangements or NPM1 mutations, Who Are Ineligible for Intensive Chemotherapy

  Transplantation and Cellular Therapy · 2026-02-01

  article
  PublisherDOI

• A study within a trial evaluating supplemental, reinforced risk-based training to reduce implementation barriers in a pediatric multicenter randomized controlled trial

  Contemporary Clinical Trials · 2026-03-13

  articleOpen access

  BACKGROUND: High-quality conduct of clinical trials depends on strict adherence to both regulatory protocols and trial-specific procedures to ensure data integrity and participant safety. Standard methods utilized to train research staff have limitations, including lack of specificity and declining impact over time. We developed a Supplemental, Reinforced, Risk-Based Training (SRRBT) program, designed to target study-specific challenges and provide ongoing reinforcement. Our objective was to evaluate whether SRRBT improves protocol adherence. METHODS: SRRBT was executed as a cluster-randomized, educational implementation study embedded in a multicenter clinical trial. We compared standard training with and without the addition of SRRBT across 23 clinical sites. SRRBT consisted of five interactive electronic vignettes focusing on trial-specific competencies. The primary outcome was the proportion of eligible participants approached for study participation. Secondary outcomes included consent and enrollment rates, protocol deviations, and completeness of study documentation. RESULTS: Approach rates immediately following training trended higher at SRRBT sites compared to those receiving standard training alone (OR = 3.98, 95% CI [0.87, 18.17], p = 0.074); however, this difference was not statistically significant and attenuated over time, with odds ratios of 2.24 (95% CI [0.78, 6.49], p = 0.14) at six months and 1.38 (95% CI [0.53, 3.59], p = 0.51) at 12 months post-training. CONCLUSIONS: SRRBT shows potential in enhancing early site-level protocol adherence but may require ongoing reinforcement to sustain benefits over time. This study highlights the need for continuous, adaptive training approaches in complex clinical trials to maintain high standards of protocol adherence.

  PublisherDOI

• Revised Recommendations for Restarting Teclistamab Following Dose Delays: Insights from the MajesTEC-1 Study on Clinical Safety, and from Simulated Pharmacokinetics and Cytokine Dynamics

  Targeted Oncology · 2026-03-26

  articleOpen access

  Teclistamab is a B cell maturation antigen × CD3 bispecific antibody approved for relapsed/refractory multiple myeloma. Two step-up doses (SUDs) are used to mitigate the risk of cytokine release syndrome (CRS). For patients who experience dose delays, it is uncertain what length of delay necessitates repeat SUDs. We used modeling simulations and retrospective analysis of the phase 1/2 MajesTEC-1 study to optimize recommendations for repeat SUDs after teclistamab dose delay. Population pharmacokinetic modeling was used to simulate teclistamab serum concentrations after dose delays to assess the duration required to achieve levels comparable to estimated trough concentrations (Ctrough) following SUDs. Quantitative systems pharmacology modeling was used to simulate cytokine dynamics. Modeling-informed time windows were applied to a retrospective analysis of CRS data from MajesTEC-1 recommended phase 2 dose cohorts to further evaluate CRS incidence with prolonged dose delays (> 28 days). Median teclistamab serum concentrations were estimated to drop to levels comparable to the simulated SUD 2 median Ctrough after 62 days and SUD 1 median Ctrough after 111 days. Simulated cytokine peaks at treatment restart during weekly or biweekly dosing at these intervals were lower than those following the initial SUD. Retrospective analysis of clinical data revealed a low incidence of CRS (grade 1–2; 2/61 [3.3%]) upon resuming teclistamab as recommended following a dose delay. These analyses suggest teclistamab can be safely restarted without repeat SUD for delays ≤ 62 days, and at SUD 2 for delays 63–111 days. For delays > 111 days, both SUDs should be administered as instructed per label before resuming treatment at 1.5 mg/kg. NCT03145181 (phase 1, May 9, 2017); NCT04557098 (phase 2, September 21, 2020). Teclistamab is used to treat patients with multiple myeloma, a type of blood cancer, that has returned after prior treatment (relapsed) or that does not respond to other treatments (refractory). Patients treated with teclistamab first receive two lower doses (step-up doses) to reduce the risk of side effects, including cytokine release syndrome. Cytokine release syndrome is caused by the rapid release of immune-signaling proteins called cytokines into the bloodstream and can result in fever, low blood pressure, and low blood oxygen levels. After step-up dosing, patients are dosed with 1.5 mg/kg once weekly. Patients who achieve a complete response or better and maintain it for at least 6 months can switch to 1.5 mg/kg every other week dosing. Some patients may need to delay teclistamab dosing to help manage side effects or due to other circumstances. We used mathematical model-based simulations to estimate how dose delays would affect serum levels of teclistamab and cytokines; these models suggested that, following dose delays, teclistamab serum levels would be comparable to, and cytokine levels lower than, the levels estimated after step-up doses. Importantly, data from the MajesTEC-1 clinical trial showed that only two of the 61 patients with dosing delays of more than 28 days experienced cytokine release syndrome; all events were low severity. Overall, modeling results and clinical trial data supported an update of the recommendations for restarting teclistamab.

  PublisherOA PDFDOI

• Abstract 4366548: Ferumoxytol Pharmacokinetics in Cardiac Magnetic Resonance

  Circulation · 2025-11-03

  article

  Introduction: Cardiac MRI (CMR) provides a non-invasive, radiation sparing assessment of anatomy and hemodynamics and is essential in caring for patients with congenital heart disease (CHD). Ferumoxytol (AMAG Pharmaceuticals) is an intravenously administered iron oxide nanoparticle that is FDA approved for the treatment of iron deficiency anemia but is increasingly used as an off-label MRI contrast agent. There are few studies assessing the effect of an iron-based contrast agent on circulating iron and iron stores in pediatric patients. Furthermore, patients with CHD may be cyanotic or have compensatory polycythemia, leading to increased iron burden. The objective of this study is to investigate the effect of radiologic dosing of Ferumoxytol on the iron stores of pediatric patients with CHD. Methods: A prospective observational study was performed on patients < 26 years of age undergoing ferumoxytol enhanced CMR (FE-CMR) at a large tertiary care center. Ferumoxytol dose was 4mg/kg administered intravenously over 15 minutes. Baseline evaluation of serum iron, transferrin, total iron binding capacity, ferritin and complete blood count were obtained within 24 hours prior to Ferumoxytol administration. Serial laboratory evaluations were performed at 3 follow up timepoints: 2-4 hours, 2-15 days, and 4-6 weeks post-contrast. Results: A total of 17 patients were enrolled. Two patients had low serum iron levels prior to FE-CMR while 15 patients had normal levels. Serum iron levels increased 2-4 hours after contrast but returned to normal or low normal levels in 2-9 days. Ferritin levels were not affected by ferumoxytol infusion. Transferrin levels and TIBC remained stable or decreased after ferumoxytol infusion. Average platelet levels before infusion were 281K/uL. No patients had an adverse reaction to infusion or clinical concern for iron toxicity. Conclusion: In pediatric patients with CHD, there is a marked increase in serum iron immediately after ferumoxytol infusion with return to baseline levels after several days. Ferumoxytol does not affect Ferritin levels. Transferrin levels and TIBC are normal or mildly decreased after infusion. This data suggests patients undergoing FE-CMR experience a transient increase in iron levels but no change to long term whole body iron levels. Further studies are needed to determine if these trends are statistically significant in a larger cohort.

  PublisherDOI

• Severe intracranial hemorrhage or ischemia associated with unfavorable outcomes after pediatric in-hospital cardiac arrest

  Resuscitation · 2025-11-08 · 1 citations

  articleOpen accessSenior author
  PublisherOA PDFDOI

• Long-Term Follow-Up from MajesTEC-1 China Cohort of Teclistamab in Patients with Relapsed/Refractory Multiple Myeloma: Efficacy Updates, Infection Profile and Immunoglobulin Usage

  Clinical Lymphoma Myeloma & Leukemia · 2025-09-01

  article
  PublisherDOI

• Multiple Model Optimal Sampling Promotes Accurate Vancomycin Area-Under-the-Curve Estimation Using a Single Sample in Critically Ill Children

  Therapeutic Drug Monitoring · 2025-01-23 · 2 citations

  articleOpen access

  BACKGROUND: Area-under-the-curve (AUC)-directed vancomycin therapy is recommended; however, AUC estimation in critically ill children is difficult owing to the need for multiple samples and lack of informative models. METHODS: The authors prospectively enrolled critically ill children receiving intravenous (IV) vancomycin for suspected infection and evaluated the accuracy of Bayesian estimation of AUC from a single, optimally timed sample. During the dosing interval, when clinical therapeutic drug monitoring was performed, an optimally timed sample was collected, which was determined for each subject using an established population pharmacokinetic model and the multiple model optimal function of Pmetrics, a nonparametric population pharmacokinetic modeling software. The model was embedded in InsightRx NOVA (InsightRx, Inc.) for individual Bayesian estimation of AUC using the optimal sample versus all available samples (optimally timed sample + clinical samples). RESULTS: Eighteen children were included. The optimal sampling time to inform Bayesian estimation of vancomycin AUC was highly variable, with trough samples being optimally informative in 32% of children. Optimal samples were collected by clinical nurses within 15 minutes of the goal time in 14 of 18 participants (78%). Compared with all samples, Bayesian AUC estimation with optimal samples had a mean bias of 0.4% (±5.9%) and mean imprecision of 4.6% (±3.6%). Bias of optimal sampling was <10% for 17 of the 18 participants (94%). When estimating AUC using only a peak sample (≤2 hours after dose) or only a trough (≤30 minutes before next dose), bias was <10% for 78% and 86% of participants, respectively. CONCLUSIONS: Optimal sampling supports accurate Bayesian estimation of vancomycin AUC from a single plasma sample in critically ill children.

  PublisherOA PDFDOI

• Urinary Biomarkers and Attainment of Cefepime Therapeutic Targets in Critically Ill Children

  The Pediatric Infectious Disease Journal · 2025-03-07 · 1 citations

  articleOpen accessSenior authorCorresponding

  BACKGROUND: Critically ill children are at risk for subtherapeutic antibiotic concentrations. The frequency of target attainment and risk factors for subtherapeutic concentrations of cefepime in children have not been extensively studied. METHODS: We performed an observational study in critically ill children receiving a new prescription of standard dosing of cefepime for suspected sepsis (≥2 systemic inflammatory response syndrome criteria within 48 hours of cefepime start). Three plasma cefepime concentrations were measured at steady state and, a urine sample was collected prior to pharmacokinetics (PK) sampling for measurement of urinary biomarkers. Bayesian analysis determined cefepime PK for each individual, and simulations were used to estimate time above minimum inhibitory concentration ( f T > MIC) for 8 µg/mL (breakpoint for Pseudomonas ). Clinical factors and urinary biomarkers were compared between patients who did and did not achieve 100% f T > MIC. Correlations between covariates and cefepime PK parameters, as well as optimal cut points to identify <100% f T > MIC, were evaluated. RESULTS: Twenty-one subjects were enrolled and PK sampling occurred after a median of 5 doses (range, 3-9); 43% of children achieved 100% f T > MIC for an MIC of 8 µg/mL. Younger age and lower urinary biomarkers (neutrophil gelatinase-associated lipocalin and kidney injury molecule-1) were significantly associated with failure to attain 100% f T > 8 µg/mL. Urinary neutrophil gelatinase-associated lipocalin (<122.1-ng/mg creatinine) best identified individuals who failed to attain this putative target (positive predictive value, 91.7%). CONCLUSIONS: A large proportion of critically ill children failed to attain target concentrations for empiric treatment of Pseudomonas aeruginosa with cefepime. Urinary biomarkers may be a noninvasive means to identify those at higher risk for increased cefepime clearance and subtherapeutic concentrations.

  PublisherOA PDFDOI

• Tranexamic acid dosing in pediatric trauma: Dose simulation based on population pharmacokinetic modeling in adult trauma patients

  Transfusion · 2025-12-23

  articleOpen access

  Abstract Background Trauma is the most common cause of death in children &gt;1 year of age, with hemorrhage being the most common preventable cause of death after injury. Antifibrinolytics like tranexamic acid (TXA) are a key aspect of trauma management in children, but optimal dosing remains unknown. The objective of this study was to derive a TXA dose in children with trauma‐related bleeding that approximates the TXA exposure in adult trauma patients using a population pharmacokinetic (popPK) model from adults with severe traumatic injury. Study Design and Methods Model‐based simulation was performed by extrapolating a previously published popPK model of TXA in adults with trauma‐related bleeding to pediatric patients. A virtual pediatric trauma population was simulated utilizing published covariate values and an allometrically scaled adult model applied to predict the TXA PK profile in children with trauma‐related bleeding. Results An IV TXA bolus of 25 mg/kg (max 2 g) in children with trauma‐related bleeding approximates the C max after a 2 g IV bolus in an adult trauma population. No dose from 20 to 35 mg/kg achieved the AUC 0–4h or AUC 0–8h that results from a 2 g IV bolus in adults. Discussion In children with trauma‐related bleeding, a TXA 25 mg/kg IV bolus is predicted to approximate the C max achieved with a 2 g IV bolus in adults. More frequent dosing may be necessary in children to achieve a similar total drug exposure as adults with trauma‐related bleeding.

  PublisherOA PDFDOI

• Association of Early Epinephrine with Hemodynamics and Outcome in Pediatric in-Hospital Cardiac Arrest: A Secondary Analysis of a Multicenter, Cluster-randomized Clinical Trial Intensive Care Unit Resuscitation (ICU-RESUS)

  Annals of the American Thoracic Society · 2025-06-04 · 1 citations

  articleOpen access

  Abstract Rationale Delayed (&amp;gt;5 minutes) epinephrine during pediatric in-hospital cardiac arrest (IHCA) is associated with worse outcomes. Epinephrine is nearly always given earlier, limiting 5 minutes as a quality target. Objectives To assess early epinephrine administration (⩽2 minutes) on outcomes and hemodynamics during cardiopulmonary resuscitation (CPR) in pediatric IHCA from pulseless, nonshockable rhythms. Methods This study leveraged the database of the ICU-RESUS (Intensive Care Unit Resuscitation) project (clinicaltrials.gov identifier NCT 02837497). Primary exposure was the time to epinephrine bolus: early versus &amp;gt;2 minutes. Primary outcome was survival to discharge. Secondary outcomes included the return of spontaneous circulation (ROSC), survival with favorable neurologic outcome, change from baseline to discharge Functional Status Scale (FSS) score, total FSS score at discharge, new morbidity among survivors, and invasively measured blood pressure during the first 10 minutes of CPR. Results Among 352 CPR events, median age was 1.0 (interquartile range [IQR], 0.3–8.0) year, 186 (53%) were male, and 185 (52.6%) had cardiac disease. Early epinephrine was administered in 273 (78%), and median time to administration was 1.0 (0.0–2.0) minute. Survival to discharge was similar between patients who received early epinephrine and those who did not. Early epinephrine administration was associated with higher ROSC, a change from baseline to discharge in FSS, lower total FSS scores at discharge, and lower rates of new morbidity compared with epinephrine administration at &amp;gt;2 minutes. The probability of ROSC and survival to discharge with favorable neurologic outcome decreased for each minute of delay in epinephrine administration. There was no difference in the invasive blood pressure targets during the first 10 minutes of CPR. Conclusions Early epinephrine administration was common and was associated with higher ROSC and improved functional outcomes compared with epinephrine administration at &amp;gt;2 minutes in pediatric IHCA.

  PublisherOA PDFDOI

Recent grants

• NIH Grant M01RR000240

  NIH · $6.4M · 2006

• Pediatric Critical Care Research Network at Children's Hospital of Philadelphia

  NIH · $1.7M · 2009–2020

• Impact of Pharmacology on Duration of Ventilation in Patients with Resp Failure

  NIH · $1.7M · 2009–2014

• RESTORE resilience in critically ill children - R2

  NIH · $506k · 2017–2021

• NIH Grant K23HD048817

  NIH · $662k · 2010

Frequent coauthors

• Kathleen L. Meert

  318 shared

• Robert A. Berg

  University of Pennsylvania

  307 shared

• Joseph A. Carcillo

  University of Pittsburgh

  305 shared

• Peter M. Mourani

  Arkansas Children's Hospital

  273 shared

• Anil Sapru

  University of California, Los Angeles

  258 shared

• Andrew R. Yates

  Nationwide Children's Hospital

  237 shared

• Ron Reeder

  University of Utah

  208 shared

• Daniel A. Notterman

  Princeton University

  198 shared

Labs

• Athena F. Zuppa LaboratoryPI