Charu Aggarwal
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Charu Aggarwal has a total of 459 co-authors. Here are the top co-authors by number of shared publications:
- Joel W. Neal
Stanford University, Stanford, CA
32 shared publications - Dylan Scholes
Division of Hematology-Oncology, Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA
30 shared publications - Sarah B. Goldberg
Department of Medicine (Medical Oncology), Yale School of Medicine, New Haven, Connecticut
30 shared publications - Andrew J. Piper Vallillo
Department of Medicine, Division of Hematology/Oncology, Lahey Hospital and Medical Center, Burlington, MA, United States
29 shared publications - Anna Wurtz
Department of Medicine (Medical Oncology), Yale University School of Medicine, New Haven, Connecticut
29 shared publications - Christina J. Falcon
Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA
29 shared publications
Latest Publications
- Open Access
A Computational Histology Artificial Intelligence Prognostic Biomarker in Non-Small Cell Lung Cancer Using the Cancer Genome Atlas
Background/Objectives: Outcomes in non–small cell lung cancer (NSCLC) remain heterogeneous, even within stage and molecular subtypes. We evaluated whether a Computational Histology Artificial Intelligence (CHAI) biomarker, derived solely from the diagnostic hematoxylin-and-eosin (H&E) whole-slide image (WSI), provides prognostic information independent of established clinicopathologic factors. Methods: Using The Cancer Genome Atlas (TCGA) lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) projects, 914 patients with an evaluable diagnostic digitized whole-slide images (WSI) and outcomes data were stratified by stage and histology and randomly split into a development set (30%, N = 270) and a held-out validation set (70%, N = 644). A CHAI histologic signature was developed and locked on the development set and applied without modification to the validation set. In development, image features were extracted from H&E-stained WSIs and used to train a Cox proportional hazards model to output a CHAI prognostic biomarker score and then dichotomized into high (CHAI positive (+))- and low (CHAI negative (−))-risk groups. The primary endpoint was overall survival (OS), and the secondary endpoint was progression-free interval (PFI). Associations were assessed by Kaplan–Meier and Cox proportional-hazards models and adjusted for traditional clinicopathologic and genomic variables. Results: A total of 644 patients were included in the validation cohort. Median age was 68 (IQR: 60–74), 388 (60%) were male, 317 (49%) had LUAD, 327 (51%) had LUSC, and 57 (9%) were never smokers. CHAI-positive patients had significantly worse OS than CHAI-negative patients (3-year OS: 53% vs. 69%; log-rank p < 0.001). The biomarker remained significantly associated with OS (HR 1.70, 95% CI 1.29–2.24, p < 0.001) and PFI (HR 1.39, 95% CI 1.05–1.84, p = 0.023) in multivariable analysis. The effect was consistent across both histologies, and a model combining the CHAI score with clinicopathologic variables was well-calibrated in validation. The biomarker was not associated with stage, age, sex, or histologic subtype, and its adjusted prognostic effect was unchanged after relevant genomic alterations were added to the model. CHAI-positive tumors were, however, enriched for KRAS mutations (21% vs. 11% wild-type, p adjusted = 0.005). Conclusions: An H&E-only CHAI NSCLC biomarker was developed and provided independent prognostic stratification in a held-out NSCLC validation cohort. Such a biomarker has the potential to refine risk stratification in NSCLC.
Open AccessA Computational Histology Artificial Intelligence Prognostic Biomarker in Non-Small Cell Lung Cancer Using the Cancer Genome Atlas
Background/Objectives: Outcomes in non–small cell lung cancer (NSCLC) remain heterogeneous, even within stage and molecular subtypes. We evaluated whether a Computational Histology Artificial Intelligence (CHAI) biomarker, derived solely from the diagnostic hematoxylin-and-eosin (H&E) whole-slide image (WSI), provides prognostic information independent of established clinicopathologic factors. Methods: Using The Cancer Genome Atlas (TCGA) lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) projects, 914 patients with an evaluable diagnostic digitized whole-slide images (WSI) and outcomes data were stratified by stage and histology and randomly split into a development set (30%, N = 270) and a held-out validation set (70%, N = 644). A CHAI histologic signature was developed and locked on the development set and applied without modification to the validation set. In development, image features were extracted from H&E-stained WSIs and used to train a Cox proportional hazards model to output a CHAI prognostic biomarker score and then dichotomized into high (CHAI positive (+))- and low (CHAI negative (−))-risk groups. The primary endpoint was overall survival (OS), and the secondary endpoint was progression-free interval (PFI). Associations were assessed by Kaplan–Meier and Cox proportional-hazards models and adjusted for traditional clinicopathologic and genomic variables. Results: A total of 644 patients were included in the validation cohort. Median age was 68 (IQR: 60–74), 388 (60%) were male, 317 (49%) had LUAD, 327 (51%) had LUSC, and 57 (9%) were never smokers. CHAI-positive patients had significantly worse OS than CHAI-negative patients (3-year OS: 53% vs. 69%; log-rank p < 0.001). The biomarker remained significantly associated with OS (HR 1.70, 95% CI 1.29–2.24, p < 0.001) and PFI (HR 1.39, 95% CI 1.05–1.84, p = 0.023) in multivariable analysis. The effect was consistent across both histologies, and a model combining the CHAI score with clinicopathologic variables was well-calibrated in validation. The biomarker was not associated with stage, age, sex, or histologic subtype, and its adjusted prognostic effect was unchanged after relevant genomic alterations were added to the model. CHAI-positive tumors were, however, enriched for KRAS mutations (21% vs. 11% wild-type, p adjusted = 0.005). Conclusions: An H&E-only CHAI NSCLC biomarker was developed and provided independent prognostic stratification in a held-out NSCLC validation cohort. Such a biomarker has the potential to refine risk stratification in NSCLC.
- Open Access
The Waiting Game: How NGS and Guideline-Concordant-Care Timing Shape Survival in Advanced NSCLC
Background: Next-generation sequencing (NGS) and the initiation of biomarker-driven first-line (1L) treatment on overall survival (OS) was evaluated in advanced/metastatic non-small-cell lung cancer (a/mNSCLC). By applying a time-dependent regression modeling approach accounting for fluidity of clinical care, this study aimed to capture the dynamic interplay between NGS testing, turnaround time (TaT), treatment decisions and timing of initiation, and outcomes more comprehensively than traditional static analyses. Methods: This study was a retrospective observational study using the US Flatiron Health EHR-derived de-identified database. Eligible adults had non-squamous a/mNSCLC and received blood-based (B-NGS) or tissue-based (T-NGS) NGS within 90 days of diagnosis (1 January 2018–30 June 2023). A time-dependent covariate captured the NGS TaT, presence of a targetable alteration, time to 1L initiation, and 1L regimen (targeted concordant vs. discordant). OS was measured from the NGS order date to death. Results: A total of 1806 patients underwent B-NGS and 2583 underwent T-NGS. Starting empiric discordant 1L therapy before NGS results was associated with higher mortality versus initiating targeted concordant therapy after biomarker identification: B-NGS hazard ratio (HR): 1.58 (95% CI, 1.10–2.26); T-NGS HR: 1.07 (0.82–1.39). Considering discordant 1L therapy irrespective of timing, mortality remained higher versus targeted concordant therapy: B-NGS HR: 1.35 (1.04–1.74); T-NGS HR: 1.24 (0.98–1.56). Conclusions: This study demonstrates that waiting for comprehensive molecular profiling results before starting 1L treatment enables the delivery of appropriate biomarker-driven therapy, which translates into improved patient outcomes. Delays in NGS result availability or treatment initiation were linked to worse survival, underscoring the need for streamlined diagnostic workflows and rapid testing modalities.
Open AccessThe Waiting Game: How NGS and Guideline-Concordant-Care Timing Shape Survival in Advanced NSCLC
Background: Next-generation sequencing (NGS) and the initiation of biomarker-driven first-line (1L) treatment on overall survival (OS) was evaluated in advanced/metastatic non-small-cell lung cancer (a/mNSCLC). By applying a time-dependent regression modeling approach accounting for fluidity of clinical care, this study aimed to capture the dynamic interplay between NGS testing, turnaround time (TaT), treatment decisions and timing of initiation, and outcomes more comprehensively than traditional static analyses. Methods: This study was a retrospective observational study using the US Flatiron Health EHR-derived de-identified database. Eligible adults had non-squamous a/mNSCLC and received blood-based (B-NGS) or tissue-based (T-NGS) NGS within 90 days of diagnosis (1 January 2018–30 June 2023). A time-dependent covariate captured the NGS TaT, presence of a targetable alteration, time to 1L initiation, and 1L regimen (targeted concordant vs. discordant). OS was measured from the NGS order date to death. Results: A total of 1806 patients underwent B-NGS and 2583 underwent T-NGS. Starting empiric discordant 1L therapy before NGS results was associated with higher mortality versus initiating targeted concordant therapy after biomarker identification: B-NGS hazard ratio (HR): 1.58 (95% CI, 1.10–2.26); T-NGS HR: 1.07 (0.82–1.39). Considering discordant 1L therapy irrespective of timing, mortality remained higher versus targeted concordant therapy: B-NGS HR: 1.35 (1.04–1.74); T-NGS HR: 1.24 (0.98–1.56). Conclusions: This study demonstrates that waiting for comprehensive molecular profiling results before starting 1L treatment enables the delivery of appropriate biomarker-driven therapy, which translates into improved patient outcomes. Delays in NGS result availability or treatment initiation were linked to worse survival, underscoring the need for streamlined diagnostic workflows and rapid testing modalities.
- 15 July 2026
- Published by American Society of Clinical Oncology (ASCO) in JCO Precision Oncology
- Vol. 10 (7) , e2501207
EMBER-Lung: Electronic Medical Record Boosting Molecular Testing in Early-Stage Non-Small Cell Lung Cancer.
Adjuvant targeted therapies and perioperative chemoimmunotherapy have made it imperative to perform molecular testing in early-stage nonsquamous (NSq) non-small cell lung cancer (NSCLC). We present a prospective clinical trial evaluating an electronic medical record (EMR)-based nudge intervention to promote timely completion of comprehensive molecular testing (cMT) in patients with early-stage NSq NSCLC. The EMBER-Lung trial prospectively enrolled patients undergoing surgical resection. The intervention included an EMR-based nudge at the time of the first postoperative visit, prompting the clinician to accept an order for cMT if NSq NSCLC histology. Following completion of cMT, if an actionable alteration was found, a reflex alert summarizing therapeutic options was sent to the patient's care team. The primary end point was the proportion of patients who underwent cMT in the real-world control and nudge intervention cohorts. Clinical characteristics and outcomes of the control and nudge intervention cohorts were compared using Pearson Chi-squared test, Fisher exact test, or Z score. Between July 2021 and November 2023, 460 patients were included (243 nudge and 217 control patients). The median age was 68.4 years, 64.3% female; 74.1% had a smoking history and mostly stage I (I/II/III; 75.9%, 14.1%, 10.0%, respectively). The proportion of patients with cMT increased in the nudge intervention (80/217, 36.9% v 197/243, 81.1%, P < .00001) across all stages (I to III). A greater proportion of patients with classical EGFR mutations and KRAS G12C mutations were detected in the nudge intervention setting (EGFR: 14/217, 6.5% v 41/243, 16.9%, P < .001; KRAS G12C: 12/217, 5.5%, v 33/243, 13.6%, P < .01). In well-matched control and nudge intervention cohorts, an EMR-based nudge after resection of early-stage NSq NSCLC was feasible and increased the proportion of patients with cMT. Future research will incorporate this earlier, at the time of biopsy, to inform neoadjuvant therapy.
- 15 July 2026
- Published by American Society of Clinical Oncology (ASCO) in JCO Precision Oncology
- Vol. 10 (7) , e2501207
EMBER-Lung: Electronic Medical Record Boosting Molecular Testing in Early-Stage Non-Small Cell Lung Cancer.
Adjuvant targeted therapies and perioperative chemoimmunotherapy have made it imperative to perform molecular testing in early-stage nonsquamous (NSq) non-small cell lung cancer (NSCLC). We present a prospective clinical trial evaluating an electronic medical record (EMR)-based nudge intervention to promote timely completion of comprehensive molecular testing (cMT) in patients with early-stage NSq NSCLC. The EMBER-Lung trial prospectively enrolled patients undergoing surgical resection. The intervention included an EMR-based nudge at the time of the first postoperative visit, prompting the clinician to accept an order for cMT if NSq NSCLC histology. Following completion of cMT, if an actionable alteration was found, a reflex alert summarizing therapeutic options was sent to the patient's care team. The primary end point was the proportion of patients who underwent cMT in the real-world control and nudge intervention cohorts. Clinical characteristics and outcomes of the control and nudge intervention cohorts were compared using Pearson Chi-squared test, Fisher exact test, or Z score. Between July 2021 and November 2023, 460 patients were included (243 nudge and 217 control patients). The median age was 68.4 years, 64.3% female; 74.1% had a smoking history and mostly stage I (I/II/III; 75.9%, 14.1%, 10.0%, respectively). The proportion of patients with cMT increased in the nudge intervention (80/217, 36.9% v 197/243, 81.1%, P < .00001) across all stages (I to III). A greater proportion of patients with classical EGFR mutations and KRAS G12C mutations were detected in the nudge intervention setting (EGFR: 14/217, 6.5% v 41/243, 16.9%, P < .001; KRAS G12C: 12/217, 5.5%, v 33/243, 13.6%, P < .01). In well-matched control and nudge intervention cohorts, an EMR-based nudge after resection of early-stage NSq NSCLC was feasible and increased the proportion of patients with cMT. Future research will incorporate this earlier, at the time of biopsy, to inform neoadjuvant therapy.
- Open Access
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S1 from Efficacy of Osimertinib in Patients with Lung Cancer Positive for Uncommon EGFR Exon 19 Deletion Mutations
Supplementary Figure S1. Multi-institution cohort of patients with tumors harboring EGFR Exon 19 deletions. Primary analyses were performed on patients with tumors harboring E746_A750del and L747_A750>P, while investigation into additional uncommon variants was descriptive only. Patient numbers do not reflect the true prevalence of each variant since this cohort is selectively enriched for uncommon exon 19 deletions.
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S1 from Efficacy of Osimertinib in Patients with Lung Cancer Positive for Uncommon EGFR Exon 19 Deletion Mutations
Supplementary Figure S1. Multi-institution cohort of patients with tumors harboring EGFR Exon 19 deletions. Primary analyses were performed on patients with tumors harboring E746_A750del and L747_A750>P, while investigation into additional uncommon variants was descriptive only. Patient numbers do not reflect the true prevalence of each variant since this cohort is selectively enriched for uncommon exon 19 deletions.
- Open Access
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S2 from Efficacy of Osimertinib in Patients with Lung Cancer Positive for Uncommon EGFR Exon 19 Deletion Mutations
Supplementary Figure S2. Progression Free Survival for patients with tumors harboring select exon 19 deletions treated with first-line osimertinib. The point estimate for median progression free survival associated with each mutation is listed to the right of the figure legend.
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S2 from Efficacy of Osimertinib in Patients with Lung Cancer Positive for Uncommon EGFR Exon 19 Deletion Mutations
Supplementary Figure S2. Progression Free Survival for patients with tumors harboring select exon 19 deletions treated with first-line osimertinib. The point estimate for median progression free survival associated with each mutation is listed to the right of the figure legend.
- Open Access
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Table S1 from Efficacy of Osimertinib in Patients with Lung Cancer Positive for Uncommon EGFR Exon 19 Deletion Mutations
Supplementary Table S1. Descriptive analysis of patient characteristics and outcomes for those with tumors harboring uncommon exon 19 deletions treated with osimertinib. Each variant represented ≤2.5% of the study cohort. Patients with + for “Smoking History” are those with either a prior or current smoking history. PFS=Progression free survival; OS= Overall survival; à denotes patients who have not yet experienced an event for a time to event endpoint.
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Table S1 from Efficacy of Osimertinib in Patients with Lung Cancer Positive for Uncommon EGFR Exon 19 Deletion Mutations
Supplementary Table S1. Descriptive analysis of patient characteristics and outcomes for those with tumors harboring uncommon exon 19 deletions treated with osimertinib. Each variant represented ≤2.5% of the study cohort. Patients with + for “Smoking History” are those with either a prior or current smoking history. PFS=Progression free survival; OS= Overall survival; à denotes patients who have not yet experienced an event for a time to event endpoint.
- Open Access
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Supplementary Figure S3 from Efficacy of Osimertinib in Patients with Lung Cancer Positive for Uncommon EGFR Exon 19 Deletion Mutations
Supplementary Figure S3. Progression Free Survival for patients with tumors harboring select exon 19 deletions treated with second or later line osimertinib (T790M+). The point estimate for median progression free survival associated with each mutation is listed to the right of the figure legend.
- 17 June 2026
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Figure S3 from Efficacy of Osimertinib in Patients with Lung Cancer Positive for Uncommon EGFR Exon 19 Deletion Mutations
Supplementary Figure S3. Progression Free Survival for patients with tumors harboring select exon 19 deletions treated with second or later line osimertinib (T790M+). The point estimate for median progression free survival associated with each mutation is listed to the right of the figure legend.
- 1 April 2026
- Published by American Society of Clinical Oncology (ASCO) in JCO Oncology Practice
- Vol. 22 (4) , 708-716
International Medical Graduates Representation at International Oncology Conference Meetings: An Analysis of ASCO Annual Meetings
PURPOSE: Despite international medical graduates (IMGs) accounting for up to one third of all practicing oncologists in the United States, and nearly half of fellowship trainees, little is known about their contributions to presentations at international oncology meetings. Therefore, we investigated the representation of IMGs in invited faculty roles at the ASCO Annual Meeting as a representative example. METHODS: Presentation data from ASCO 2022 and 2023 Annual Meetings were obtained from program guides. We collected presentation-level and speaker-level variables of invited faculty roles and original research. Speakers were categorized as American medical graduate (AMG), IMG, non–US-based, or non-MD (Doctor of Medicine). Univariable and multivariable analyses assessed factors influencing selection for invited faculty roles. RESULTS: A total of 2,590 presentations by 2,109 unique speakers were reviewed, including primary research presenters and invited faculty. Among the 1,352 presentations by AMGs, 30% were primary research and 70% were invited faculty roles. For the 374 presentations by IMGs, 41% were primary research and 59% were invited faculty roles. Notably, first-author IMGs were significantly less likely to receive invited faculty roles relative to their primary research output compared with first-author AMGs (adjusted odds ratio, 0.69 [95% CI, 0.53 to 0.90]; P = .006), even when accounting for academic rank, presumed sex, and year. CONCLUSION: Relative to IMGs' original research output, IMGs appear to receive fewer invitations for ASCO Annual Meeting invited faculty roles compared with AMGs. Our findings suggest potential underrepresentation and under-recognition of IMGs in these roles, despite their contributions to the oncology workforce in the United States.
- 1 April 2026
- Published by American Society of Clinical Oncology (ASCO) in JCO Oncology Practice
- Vol. 22 (4) , 708-716
International Medical Graduates Representation at International Oncology Conference Meetings: An Analysis of ASCO Annual Meetings
PURPOSE: Despite international medical graduates (IMGs) accounting for up to one third of all practicing oncologists in the United States, and nearly half of fellowship trainees, little is known about their contributions to presentations at international oncology meetings. Therefore, we investigated the representation of IMGs in invited faculty roles at the ASCO Annual Meeting as a representative example. METHODS: Presentation data from ASCO 2022 and 2023 Annual Meetings were obtained from program guides. We collected presentation-level and speaker-level variables of invited faculty roles and original research. Speakers were categorized as American medical graduate (AMG), IMG, non–US-based, or non-MD (Doctor of Medicine). Univariable and multivariable analyses assessed factors influencing selection for invited faculty roles. RESULTS: A total of 2,590 presentations by 2,109 unique speakers were reviewed, including primary research presenters and invited faculty. Among the 1,352 presentations by AMGs, 30% were primary research and 70% were invited faculty roles. For the 374 presentations by IMGs, 41% were primary research and 59% were invited faculty roles. Notably, first-author IMGs were significantly less likely to receive invited faculty roles relative to their primary research output compared with first-author AMGs (adjusted odds ratio, 0.69 [95% CI, 0.53 to 0.90]; P = .006), even when accounting for academic rank, presumed sex, and year. CONCLUSION: Relative to IMGs' original research output, IMGs appear to receive fewer invitations for ASCO Annual Meeting invited faculty roles compared with AMGs. Our findings suggest potential underrepresentation and under-recognition of IMGs in these roles, despite their contributions to the oncology workforce in the United States.
- Open Access
- 25 November 2025
- Published by American Association for Cancer Research (AACR)
Supplementary Tables S1-S3 from Safety and Efficacy of MEDI0457 plus Durvalumab in Patients with Human Papillomavirus–Associated Recurrent/Metastatic Head and Neck Squamous Cell Carcinoma
Supplementary Table S1. Representativeness of Study Participants Supplementary Table S2. Disease response as assessed by RECIST v1.1 by PD-L1 tumor cell expression (response-evaluable population). Supplementary Table S3. Disease response as assessed by RECIST v1.1 in prior line of therapy subgroups (response-evaluable population) and survival (as-treated population).
- 25 November 2025
- Published by American Association for Cancer Research (AACR)
Open AccessSupplementary Tables S1-S3 from Safety and Efficacy of MEDI0457 plus Durvalumab in Patients with Human Papillomavirus–Associated Recurrent/Metastatic Head and Neck Squamous Cell Carcinoma
Supplementary Table S1. Representativeness of Study Participants Supplementary Table S2. Disease response as assessed by RECIST v1.1 by PD-L1 tumor cell expression (response-evaluable population). Supplementary Table S3. Disease response as assessed by RECIST v1.1 in prior line of therapy subgroups (response-evaluable population) and survival (as-treated population).